Jaana Juvonen, Kathryn R. Wentzel, Bernard Weiner

How are children's social lives at school related to their motivation to
achieve and how do motivational and social processes interact to explain

children's adjustment at school? This volume, featuring work by leading
researchers in educational and developmental psychology, provides new
perspectives on how and why children tend to thrive or fail at school. The
individual chapters examine the unique roles of peers and teachers in
communicating and reinforcing school-related attitudes, expectations, and
definitions of self. Relations of children's school adjustment to school
motivation, interpersonal functioning, and social skillfulness are also ex-
plored.
The developmental and social perspectives on motivation and achieve-
ment presented in this volume provide new insights into the complex
processes contributing to school success. This book will be vital reading
for educators and developmental psychologists interested in the contextu-
al as well as intrapersonal processes that contribute to social and academ-
ic competence.
Social motivation
Understanding children's
school adjustment
Cambridge Studies in Social and Emotional Development
General Editor: Martin L. Hoffman, New York University
Advisory Board: Robert N. Emde, Willard W. Hartup,

Robert A. Hinde, Lois W. Hoffman, Carroll E. Izard,
Nicholas Blurton Jones, Jerome Kagan, Franz J. Monks,
Paul Mussen, Ross D. Parke, and Michael Rutter
Social motivation
Social motivation
Understanding children's
school adjustment
Edited by
Jaana Juvonen
University of California at Los Angeles
KathrynR.Wentzel
University of Maryland
CAMBRIDGE
UNIVERSITY PRESS
Published by the Press Syndicate of the University of Cambridge
The Pitt Building, Trumpington Street, Cambridge CB2 1RP
40 West 20th Street, New York, NY 10011-4211, USA
10 Stamford Road, Oakleigh, Melbourne 3166, Australia
© Cambridge University Press 1996
First published 1996
Library of Congress Cataloging-iti-Pubiication Data
Social motivation : understanding children's school adjustment / Jaana

Juvonen & Kathryn R. Wentzel, eds.
p. cm.
Includes bibliographical references and indexes.
ISBN 0-521-47324-1 (he). — ISBN 0-521-56442-5 (pb)
1. Achievement motivation in children. 2. Achievement motivation
in adolescence. 3. Student adjustment. 4. Social desirability in
children. 5. Social desirability in adolescence. 6. Peer pressure
in children. 7. Peer pressure in adolescence. 8. Motivation in
education. I. Juvonen, Jaana. II. Wentzel, Kathryn R.
BF723.M56S63 1996
155.42?438—dc20 95-47456
CIP
A catalog record for this book is available from the British Library
ISBN 0-521-47324-1 (he)

ISBN 0-521-56442-5 (pb)
Transferred to digital printing 2003

To our parents,
Ulla-Maija and Jaakko Juvonen,
Ruth and Herman Wentzel
Contents
Contributors page ix
Preface xi
Foreword xiii
Bernard Weiner
1 Introduction: New perspectives on motivation at school 1

Kathryn R. Wentzel
Part I Social motivation: Perspectives on self

2 Teacher and classmate influences on scholastic motivation,
self-esteem, and level of voice in adolescents 11
Susan Barter
3 Self-presentation tactics promoting teacher and
peer approval: The function of excuses and other
clever explanations 43
Jaana Juvonen
4 Social self-discrepancy: A theory relating peer relations
problems and school maladjustment 66
Janis B. Kupersmidt, Kathy S. Buchele, Mary Ellen Voegler,
and Constantine Sedikides
5 Motivational approaches to aggression within the context
of peer relationships 98
Cynthia A. Erdley
6 Motivational opportunities and obstacles associated with
social responsibility and caring behavior in school contexts 126
Martin E. Ford
VII
viii Contents
1 Modeling and self-efficacy influences on children's

development of self-regulation 154
Dale H. Schunk and Barry J. Zimmerman
8 Social motivation: Goals and social-cognitive processes.
A comment 181
Carol S. Dweck
Part II Social motivation: Perspectives on relationships

9 Interpersonal relationships in the school environment
and children's early school adjustment: The role of
teachers and peers 199
Sondra K Birch and Gary W. Ladd
10 Social goals and social relationships as motivators of
school adjustment 226
Kathryn R. Wentzel
11 Friends' influence on school adjustment: A motivational
analysis 248
Thomas J. Berndt and Keunho Keefe
12 Peer networks and students' classroom engagement during
childhood and adolescence 279
Thomas A. Kindermann, Tanya L. McCollam, and
Ellsworth Gibson Jr.
13 Academic failure and school dropout: The influence
of peers 313
Shelley Hymel, Colin Comfort, Kimberly Schonert-Reichl,
and Patricia McDougall
14 What's "emotional" about social motivation? A comment 346
Sandra Graham
Author index 361
Subject index 371

Contributors
Thomas J. Berndt Martin E. Ford

Department of Psychological Sciences Graduate School of Education
Purdue University George Mason University
West Lafayette, IN 47907 4400 University Drive
Fairfax, VA 22030
Sondra H. Birch
Department of Educational Psychology
Ellsworth Gibson, Jr.
University of Illinois
Department of Psychology
1310 S Sixth Street
Portland State University
Champaign, IL 61820
P.O. Box 751
Kathy S. Buchele Portland, OR 97207
Department of Psychology, CB#3270
University of North Carolina Sandra Graham
Chapel Hill, NC 27599 Graduate School of Education
UCLA
Colin Comfort 405 Hilgard Ave.
Faculty of Education Los Angeles, CA 90095
University of British Columbia
2125 Main Mall Susan Harter
Vancouver BC, Canada 6VT 1Z4 Department of Psychology
University of Denver
Carol S. Dweck
2155 South Race St.
Department of Psychology
Denver, CO 80208
Columbia University
119th St. and Amsterdam Ave.
New York, NY 10027 Shelly Hymel
Faculty of Education
Cynthia A. Erdley University of British Columbia
Department of Psychology 2125 Main Mall
University of Maine Vancouver BC, Canada
Orono, ME 04469 6VT 1Z4
IX
Contributors
Jaana Juvonen Kimberly Schonert-Reichl

Department of Psychology Faculty of Education
UCLA University of British Columbia
405HilgardAve. 2125 Main Mall
Los Angeles, CA 90095 Vancouver BC, Canada 6VT 1Z4
Keunho Keefe Dale H. Schunk

Reiss-Davis Child Study Center 1446 LAEB Room 5108
3200 Motor Ave. School of Education
Los Angeles, CA 90034 Purdue University
West Lafayette, IN 47907
Thomas A. Kindermann
Department of Psychology Constantine Sedikides
Portland State University Department of Psychology, CB#3270
P.O. Box 751 University of North Carolina
Portland, OR 97207 Chapel Hill, NC 27599
Janis B. Kupersmidt Mary Ellen Voegler

Department of Psychology, CB#3270 Department of Psychology, CB#3270
University of North Carolina University of North Carolina
Chapel Hill, NC 27599 Chapel Hill, NC 27599-3270
GaryW.Ladd Bernard Weiner

Department of Educational Department of Psychology
Psychology UCLA
University of Illinois 405 Hilgard Ave.
1310 S Sixth Street Los Angeles, CA 90095
Champaign, IL 61820
KathrynR.Wentzel
Tanya L. McCollam Department of Human
Department of Psychology Development
Portland State University University of Maryland
P.O. Box 751 College Park, MD 20742
Portland, OR 97207
Barry J. Zimmerman
Patricia McDougall Graduate School
Faculty of Education City University of New York
University of British Columbia 23 West 42nd Street
2125 Main Mall New York 10036
Vancouver BC, Canada 6VT 1Z4
Preface
The idea for this book developed over the last few years, when we found
ourselves at the same conferences, assigned to the same sessions, present-
ing research on children's and adolescents' social relationships in school.
We felt that it was time to pool our efforts and invite others to join us to
write about achievement motivation and social functioning, not as two
distinct topics, but as interrelated aspects of school adjustment. We both
believed that students' acclimation to school is colored by their social ex-
periences - an aspect that had often been neglected, especially in educa-
tional research. Furthermore, motivational approaches open up new ways
to interpret interpersonal interactions and relationships in the classroom -
an observation that has not been widely acknowledged in research on so-
cial development.
While editing this book, there was an increase in interest in social in-
fluences in the school setting. We want to especially thank all our contrib-
utors for their enthusiasm. The authors were at different phases of their so-
cial-motivational thinking; some were at the planning stage, while others
were in the process of compiling further data and extending their prior
findings. We welcomed this variability in conceptualizations that quite ac-
curately reflected the current state of research in this area. Throughout the
editing process, we appreciated the encouragement and support from our
friends at the University of Delaware and the University of Maryland.
Also, we want to acknowledge the support for our own research from the
National Academy of Education Spencer Fellowship Program and the Of-
fice of Educational Research and Improvement Fellows Program. Finally,
thanks are extended to Bernard Weiner for his encouragement and support
as well as to Khanh-Van Bui for her assistance.
Jaana Juvonen
Kathryn R. Wentzel
xi
Foreword
Bernard Weiner
In the middle 1940s, David McClelland and his coworkers initiated what
might be considered the first systematic experimental research pursuing
issues in human motivation. An early goal they set for themselves was to
develop a measure of human motives in order to identify individuals dif-
fering in motive strengths. A thermometer metaphor guided this work. It
was reasoned that a "good" thermometer would register an increase when
the heat was turned on. In a psychological context, the idea was to manip-
ulate some antecedent that would produce an increase in motive strength
under arousal conditions, as opposed to contexts in which there was no
stimulating occurrence.
In achievement contexts, it was reasoned that failure would function to
galvanize achievement needs. Thus, some individuals failed at an achieve-
ment task, whereas others succeeded or were given a nonachievement-re-
lated activity to perform. Then the thermometer was thrust into the mouth
(or, in this case, in front of the eyes) of the individual to ascertain if in-
creased motivation was registered. The measuring device selected was a
projective instrument called the Thematic Apperception Test (TAT). In-
deed, stories to TAT cards did contain more achievement imagery follow-
ing failure than after success or some control experience. In this manner,
it was deemed that an appropriate instrument to assess achievement de-
sires had been developed. It therefore also followed that when persons re-
spond to the TAT under neutral or nonaroused conditions, those with high-
er scores are "walking around" aroused; that is, they are more motivated
to achieve than those exhibiting lower scores in this neutral context.
This conclusion led to hundreds of studies in the achievement domain,
followed by many others that may not have used the TAT but nonetheless
had the goal of illuminating our understanding of achievement strivings.
Indeed, achievement has been the focus of research in human motivation,
xiii
xiv Foreword
incorporating self-related affects and cognitions including pride, probabil-

ity of success, ego-orientation, and the like.
This same line of reasoning was applied to the study of affiliative mo-
tivation. It was believed that if the TAT was a good thermometer to assess
achievement needs, then it also should be able to register affiliative ten-
dencies. The question then became what antecedent, what "source of
heat," in the affiliative domain would be analogous to achievement fail-
ure; that is, what would arouse affiliative concerns. The procedure select-
ed was to gather individuals from a fraternity and have them stand before
others and be rated on their desirable and undesirable qualities. After do-
ing this, the TAT was administered and the content was scored for affilia-
tive motivation. This proved to be a source of great heat, and the TAT did
register increased affiliative concerns after this experience. The heat was
so intense, however, that persons being rated by others experienced severe
stress and anxiety, some broke into tears, and the ethics as well as the effi-
cacy of the procedure was called into question. Notice that this did not
happen following an achievement failure - that was merely a small blaze,
not a forest fire.
A number of long-lasting effects on the field of motivation resulted in
part from this early history:
1. Researchers focused their efforts on the field of achievement

motivation, partially because it lent itself to experimental ma-
nipulations.
2. Researchers drew away from affiliative motivation, partially
because it was perceived as not lending itself to experimental
study.
3. Achievement and affiliation were thought of as very distinct
motivations: their interaction and interplay were not consid-
ered.
This book begins to redress this imbalance and counter mispercep-

tions. Here we see that achievement and affiliation are intertwined. Feel-
ings of rejection, lack of support, and dissatisfaction of affiliative needs
affect achievement motivation and school performance. For example, stu-
dents lonely and isolated in the classroom are likely to give up achieve-
ment strivings, and drop out of school. In addition, one's achievements in-
fluence with whom one will affiliate. For example, students with similar
achievement strivings tend to form friendships and peer groups, and those
whose behavior is consistent with the values of their instructors are ap-
Foreword xv
proved by their teachers. Finally, peers provide students with models for
appropriate academic as well as social conduct.
It also is apparent that experimental manipulations pertinent to the
study of affiliation are quite possible. But even more evident is that proce-
dures are available that can be used in field settings that enable the study
of affiliation and social influences in real-life contexts.
I regard this book as a landmark volume that not only gives affiliative
motivation its proper role and respect, but also should be of great benefit
to researchers in the achievement field, where I believe new life-lines are
necessary. Peer and other social influences cut across motivational do-
mains, including achievement, aggression, and virtually all other motiva-
tional systems. Thus, there is potential for a general theory of motivation
that has affiliation - the establishment and maintenance of social bonds -
at its very core.
1 Introduction: New perspectives on
motivation at school
Kathryn R. Wentzel
The social worlds of children are a pervasive and influential part of their
lives at school. Each day in class, children work to maintain and establish
interpersonal relationships, they strive to develop social identities and a
sense of belongingness, they observe and model social skills and stan-
dards for performance displayed by others, and they are rewarded for be-
having in ways that are valued by teachers and peers. We also know that
children who display socially competent behavior in elementary school
are more likely to excel academically throughout their middle and high
school years than those who do not (see Wentzel, 1991).
Our purpose in bringing together the authors for this volume was to en-
rich this portrayal of how children's social and academic development are
intertwined. We asked the contributors to focus on specific ways in which
children are influenced by and motivated to achieve things social as well
as academic when they are at school. In response, the authors provided us
with a diverse set of unique perspectives on social aspects of motivation.
Collectively, however, the various perspectives bring to the forefront the
range of social outcomes that children strive to achieve at school and con-
sider ways in which these outcomes contribute to and in fact represent val-
ued aspects of school adjustment. In doing so, many of the authors em-
phasize the important role of social goals, self-referent beliefs, and social
cognitions in explanations of school-related competence. In addition, the
social motivational perspectives presented in this volume draw attention
to the need for models of school success that consider not only intraper-
sonal processes as motivators of behavior but the critical role of interper-
sonal relationships and social concerns as well.
In general, research has paid little attention to the possibility that chil-
dren's social development is related to classroom motivation and school
adjustment. Indeed, much of the recent work on motivation at school por-
trays children as striving to achieve primarily intellectual outcomes in re-
sponse to academically-related aspects of curriculum and instruction (see,
2 Introduction
e.g., Ames & Ames, 1989). The empirical work presented in this volume,
however, attests to the fact that children are motivated to achieve social as
well as academic goals at school and that indeed, school children often be-
have in direct response to their social environments. We hope, therefore,
that the work presented in this volume will provide inspiration and a chal-
lenge to continue research and theory development on social motivational
perspectives and in doing so, further understanding of children's lives at
school.
Themes of the volume

The social motivational perspectives presented in this volume reflect sever-
al broad themes. At the outset, the authors remind us that children must be
socially as well as intellectually adept if they are to be successful students.
The chapters provide multiple definitions of school adjustment, ranging
from academically-related outcomes such as attitudes, values, and motiva-
tional orientations towards school (Berndt and Keefe; Birch and Ladd; Har-
ter), task engagement (Kindermann, McCollam, and Gibson), self-regula-
tion (Schunk and Zimmerman), grades and test scores (Wentzel), and
dropping out (Hymel, Comfort, Schonert, and McDougall), to social and
personal outcomes such as the quality of interpersonal relationships with
peers (Kupersmidt, Buchele, Voeller, and Sedikes), prosocial and antisocial
behavior (Erdley; Ford; Wentzel), self-esteem (Harter), and public presen-
tations of the self (Juvonen).
In addition, ways in which students' social goals and related psycho-
logical functioning can either facilitate or hinder school adjustment are
examined. In particular, the authors highlight the social goals and motives
that underlie school adjustment, including needs for belongingness and
relatedness (Birch and Ladd; Hymel et al.), needs for social approval (Ju-
vonen), in addition to needs for admiration, self-enhancement, and verifi-
cation (Berndt and Keefe), goals to achieve prosocial and socially respon-
sible outcomes (Erdle; Ford; Wentzel), and orientations toward intrinsic
and extrinsic rewards (Harter).
Several authors also discuss how needs and goals can vary as a func-
tion of levels of difficulty and proximity (Schunk and Zimmerman), as-
pects of social integration (Ford), and the importance and type of relation-
ship children have with their peers (Kupersmidt et al.). Other aspects of
intrapsychological functioning that facilitate goal pursuit also are consid-
ered. Self-efficacy and self-regulatory skills (Erdley; Schunk and Zim-
merman), attributions (Erdley; Juvonen), agency beliefs, emotions (Ford),
Introduction 3
self perceptions (Harter), and social cognitions (Erdley; Juvonen; Schunk

and Zimmerman; Wentzel) are presented as processes that are central to
the successful accomplishment of goals and the realization of personal
needs.
Finally, interpersonal relationships with teachers and peers are de-
scribed as potentially powerful social contexts that motivate student behav-
ior. Although social relationships are considered to be the central motiva-
tional forces underlying socialization and social development (Baumeister
& Leary, 1995; Hartup, 1989; Maccoby, 1992), their role in motivating
school success (or failure) has not been studied in systematic fashion (see
work by Eccles (1993) and Epstein (1989) for noteworthy exceptions).
Therefore, in addition to focusing on intrapersonal influences such as so-
cial goals and social cognitions, the authors also focus on interpersonal re-
lationships with teachers and peers as motivatiors of school success.
The chapters in this volume present several ways in which social rela-
tionships provide school-aged children with motivational contexts for de-
velopment. Several authors explore the motivational impact of group
membership by focusing on ways that peer groups and classroom teachers
communicate and reinforce specific values and attitudes toward school,
behavior, and even definitions of the self (Berndt and Keefe; Harter; Juvo-
nen; Kindermann et al.; Kupersmidt et al.; Schunk and Zimmerman).
Dyadic relationships are also discussed, with specific mention of ways
that friendships with peers and teacher-student relationships influence so-
cial goal pursuit (Kupersmidt et al.; Wentzel), satisfy social needs (Berndt
and Keefe; Birch and Ladd), and promote healthy social-emotional and
intellectual functioning (Berndt and Keefe; Birch and Ladd; Hymel et al.;
Kupersmidt et al.; Wentzel).
Finally, several chapters in this volume illustrate how social relation-
ships influence or motivate student behavior, depending on the specific
functions they serve. For example, they provide students with emotional
support and nurturance (Birch and Ladd; Hymel et al.; Kupersmidt et al.;
Wentzel). Relationships are also portrayed as sources of information -
about the self, about the expectations and values of others, and about how
to accomplish specific tasks (Berndt and Keefe; Harter; Juvonen; Kinder-
mann et al.; Schunk and Zimmermann).
Challenges to the field

The notion that social concerns and influences are relevant for under-
standing motivation and academic performance is certainly not new. In the
4 Introduction
early 60s, the need for Affiliation, identified by McClelland (1987) as a

central motive in people's lives, was linked to academic performance in-
dependent of need for Achievement (McKeachie, 1961). Crandall's work
(1963) related classroom achievement to children's need for social ap-
proval, and Veroff (1969) proposed that social comparison was an integral
part of motivation to achieve in school-aged children.
More recently, theory and research has focused on motivational factors
embedded in learning contexts. Research on classroom reward structures
(Ames, 1984; deCharms, 1984), organizational culture and climate
(Maehr and Midgley, 1991), and person-environment fit (Eccles and
Midgley, 1989) has greatly expanded our understanding of how the social
environments within which learning takes place can motivate children to
learn and behave in very specific ways. Theories of motivation once again
recognize social and affiliative concerns as basic goals and needs that un-
derlie student achievement (Connell & Wellborn, 1991). We believe that
the perspectives presented in this volume contribute to and extend this
work by focusing on the motivational significance of soc/aZ-psychological
functioning and the more proximal contexts of interpersonal relationships.
We hope that the perspectives introduced in this volume provide the
field with new questions and challenges for understanding motivation and
adjustment at school. For instance, can models of motivation traditionally
used to understand skill and task-related performance be used to under-
stand the development of social competence? Juvonen's extension of attri-
bution theory to the domain of social information processing and social
behavior provides one example of how this can be accomplished. Berndt
and Keefe's use of expectancy-value theory to understand the role of
friendships in school adjustment provides another. Ford's Motivational
Systems Theory illustrates how general principles of motivation can be
applied to the social domain of caring.
We might also ask how theories of socialization and social develop-
ment can inform models of motivation and school adjustment. Work de-
scribed in chapters by Harter and Schunk and Zimmerman illustrates how
social role expectations and modeling can be powerful socializers of self
processes that motivate academic behavior. Chapters by Birch and Ladd,
and Wentzel suggest how family socialization processes known to pro-
mote healthy social and emotional development might also reflect mecha-
nisms by which teachers and peers motivate social competence as well as
academic excellence in the classroom.
A consideration of interpersonal relationships as motivational contexts
clearly introduces additional levels of complexity to models of motiva-
Introduction 5
tion. Using theories of social cognition and identity development, several

authors suggest that the degree to which relationships are influential de-
pends in part on the unique perspectives and belief systems of the individ-
ual (Harter), and the degree to which the individual can identify with so-
cialization agents (whether they are teachers or peers) along certain
dimensions (Juvonen; Schunk and Zimmerman). Other authors suggest
that when relationships do form, they operate as systems that have quali-
ties separate from those of the individuals that form the relationship
(Berndt and Keefe; Birch and Ladd).
When considered as a whole, the perspectives presented in this volume
suggest that further exploration of social-cognitive and other social psy-
chological factors, and identification of variables that capture the quality
of relationships and social contexts, must take place before the nature of
"social" motivation can be understood fully. Commentaries by Dweck and
Graham suggest that perhaps even more elaborate models of motivation
might be necessary to accomplish this task. Carol Dweck provides us with
a complex and dynamic model of psychological processes that underlie
accomplishments in both social and academic domains of functioning.
Using an attributional perspective, Sandra Graham highlights the role of
emotions in describing the mechanisms that link interpersonal relation-
ships to social cognitions and student behavior. Both Dweck and Graham
remind us that the search for specific mechanisms and processes that can
explain social influences on motivation and behavior is still in its incep-
tion and remains a challenge for the field. We hope that the chapters in
this volume will provide the impetus for continued efforts in this regard.
Overview of the chapters

Although there was considerable overlap in themes and perspectives
across the eleven chapters, we divided the book into two sections. The
first set of chapters focuses primarily on intrapsychological processes that
motivate school adjustment.
Susan Harter provides a developmental perspective on three aspects of
school adjustment, intrinsic/extrinsic dimensions of motivation, self-
esteem, and students' "voice." In her discussion of these outcomes, Harter
draws particular attention to the need to study change over time, individ-
ual differences in student's perceptions of and reactions to change, and the
role of teachers and peers in the social construction of the self.
Jaana Juvonen discusses school adjustment in terms of children's un-
derstanding of implicit norms for social conduct. Drawing on attribution
6 Introduction
theory, she illustrates how students use this knowledge to present them-
selves in ways that will lead to social approval from teachers as well as
peers.
Janis Kupersmidt, Kathy Buchele, Mary Ellen Voegler, and Constan-
tine Sedikides consider a range of social and academic outcomes in rela-
tion to problematic peer relationships. In particular, Kupersmidt and her
colleagues describe ways in which social needs and social-cognitive func-
tioning mediate links between peer relationship problems and school mal-
adjustment.
Cynthia Erdley focuses on aggression as a specific example of malad-
justed school behavior. In her chapter, she describes a model of social in-
formation processing that can be used to explain children's aggressive be-
havior and subsequent peer acceptance.
Martin Ford presents a model of motivation for explaining socially re-
sponsible and caring behavior at school. Motivational Systems Theory de-
scribes a set of psychological processes including goals, emotions, and
agency beliefs that motivate the achievement of these social outcomes.
Finally, Dale Schunk and Barry Zimmerman discuss the social origins
of academic self-regulation and self-efficacy. Drawing from principles of
observational learning, these authors describe a process of internalization
whereby observed academic skills become self-regulated.
The second set of chapters focuses primarily on the role of social rela-
tionships in motivating school adjustment:
Sondra Birch and Gary Ladd discuss the influence of teacher and peer
relationships on academic as well as social aspects of adjustment. These
authors describe a "relationship features" perspective that highlights the
role of specific functions of relationships in motivating student behavior.
Kathryn Wentzel defines school adjustment in terms of prosocial and
socially responsible behavior. Her work illustrates ways in which these as-
pects of social competence contribute to academic achievement and how
relationships with teachers and peers motivate children to behave in these
socially desirable ways.
Thomas Berndt and Keunho Keefe consider attitudes, classroom be-
havior, and academic achievement as aspects of school adjustment that are
influenced by friends. These authors discuss two alternative pathways by
which friends influence academic outcomes, focusing primarily on social
motives as mechanisms that link friendship quality to school adjustment.
Thomas Kindermann, Tanya McCollam, and Ellsworth Gibson consid-
er academic motivation and engagement as aspects of school adjustment.
Introduction 7
These authors describe the role of peer networks in socializing motiva-

tional orientations toward schoolwork over the course of the school year.
In the final chapter of the volume, Shelley Hymel, Colin Comfort,
Kimberly Schonert-Reichl, and Patricia McDougall document the prob-
lem of school dropouts. Hymel and her colleagues highlight the role of
peer relationships in motivating students to either stay in or drop out of
school.
References
Ames, C. (1984). Competitive, cooperative, and individualistic goal structures: A
cognitive - motivational analysis. In R. Ames & C. Ames (Eds.), Research in
motivation in education (vol.1; pp. 177-208). New York: Academic Press.
Ames, C, & Ames, R. (1989). Research on motivation in education (vol. 3). San
Diego, CA: Academic Press.
Baumeister, R. R, & Leary, M. R. (1995). The need to belong: Desire for interper-
sonal attachments as a fundamental human motivation. Psychological Bulletin,
777,497-529.
Connell, J. P., & Wellborn, J. G. (1991). Competence, autonomy, and relatedness:
A motivational analysis of self-system processes. In M. R. Gunnar & L. A.
Sroufe (Eds.), Self processes and development: The Minnesota symposia on
child development (vol. 23; pp. 43-78). Hillsdale, NJ: Erlbaum.
Crandall, V J. (1963). Achievement. In H. W. Stevenson (Ed.), Child psychology:
Sixty-second Yearbook of the National Society for the Study of Education.
Chicago: University of Chicago Press.
deCharms, R. (1984). Motivation enhancement in educational settings. In R.
Ames & C. Ames (Eds.), Research in motivation in education (vol.1; pp. 275 -
310). New York: Academic Press.
Eccles, J. S. (1993). School and family effects on the ontogeny of children's inter-
est, self-perceptions, and activity choices. In J. Jacobs (Ed.), Nebraska sympo-
sium on motivation 1992: Developmental perspectives on motivation (pp.
145-208). Lincoln, NE: University of Nebraska Press.
Eccles, J. S., & Midgley, C. (1989). Stage-environment fit: Developmentally ap-
propriate classrooms for young adolescents. In C. Ames & R. Ames (Eds.), Re-
search on motivation in education (vol.3, pp. 139-186). New York: Academic
Press.
Epstein, J. L. (1989). The selection of friends: Changes across the grades and in
different school environments. In T. Berndt & G. Ladd (Eds.), Peer relation-
ships in child development (pp. 158-187). New York: Wiley.
Hartup, W. W. (1989). Social relationships and their developmental significance.
American Psychologist, 44, 120-126.
8 Introduction
Maccoby, E. E. (1992). Trends in the study of socialization: Is there a Lewinian

heritage? Journal of Social Issues, 48, 171-185.
Maehr, M. L., & Midgley, C. (1991). Enhancing motivation: A schoolwide ap-
proach. Educational Psychologist, 26, 399^427.
McClelland, D. C. (1987). Human motivation. New York: Cambridge University
Press.
McKeachie, W. J. (1961). Motivation, teaching methods, and college learning. In
M. R. Jones (Ed.), Nebraska symposium on motivation: 1961. Lincoln, NB:
University of Nebraska Press.
Wentzel, K. R. (1991). Social competence at school: The relation between social
responsibility and academic achievement. Review of Educational Research, 61,
1-24.
Veroff, J. (1969). Social comparison and the development of achievement motiva-
tion. In C. P. Smith (Ed.), Achievement-related motives in children. New York:
Russell Sage.
PART ONE
Social motivation:
Perspectives on self
Teacher and classmate influences on
scholastic motivation, self-esteem,
and level of voice in adolescents
Susan Harter
The classroom setting represents not only an educational arena but a pow-
erful social context in which the psychological adjustment of children and
adolesents can be affected. This chapter will focus specifically on the role
played by teachers and classmates. Teachers not only instruct, but serve to
represent and communicate a particualr educational philosophy, including
the standards by which students will be evaluated. They not only provide
feedback regarding students' academic performance, but have a major im-
pact on students' motivation to learn. Not only do they convey specific ap-
proval or disapproval for scholastic achievement, but teachers communi-
cate their more general approval or disapproval for the child as a person
(see Birch & Ladd; Wentzel, this volume). Classmates serve as potential
companions and friends, meeting important social needs of the developing
child. However, they also represent a very salient social reference group
that invites intense social comparison. In addition, the approval or disap-
proval that classmates display can have a major effect on a child's or ado-
lescent's sense of self (see also Berndt & Keefe; Kindermann, et al. this
volume).
In this chapter, I will examine the impact of teachers and classmates on
three constructs that represent different indices of adjustment within the
school context. These include: (a) intrinsic and extrinsic motivation for
classroom learning, (b) self-esteem, and (c) level of "voice," namely the
ability to express one's opinions in the classroom. Specific attention will
be given to the antecedents and mediators of the constructs, leading to an
emphasis on individual differences. Such an approach forces one to ques-
tion the reality of certain generalizations that have dominated the litera-
ture, for example, that most students lose their intrinsic motivation as they
move through the school system, that self-esteem suffers in early adoles-
cence, particularly for girls, and that at this same developmental juncture,
girls' voices go underground. A consideration of the determinants of these
11
12 Perspectives on self
constructs will enhance our understanding of individual differences and

highlight the need to exercise caution in promoting or accepting such gen-
eralizations.
Intrinsic/extrinsic motivation
The distinction between intrinsic and extrinsic motivation has proved vi-
able not only as a framework for predicting and interpreting behavior, in
general (deCharms, 1968; Deci, 1975; Deci & Ryan, 1985; Lepper,
Greene, & Nisbett, 1973) but for understanding children's behavior within
the educational setting, in particular (deCharms, 1976; Condry &
Koslowski, 1979; Connell & Ryan, 1984; Eccles & Midgley, 1988, 1990;
Eccles, Midgley, & Adler, 1984; Harter, 1981, 1992; Harter & Connell,
1984). With regard to the classroom context, investigators have posed the
question of whether a given child's motivation to learn is driven by an in-
trinsic interest in the subject material, by curiosity and preference for
challenge, or by an extrinsic orientation in which one is motivated by the
desire to obtain grades, to win teacher approval, to avoid censure, namely,
to meet the external demands of the school system.
In our own work, we were initially interested in the developmental
course of these motivational constructs. The instrument that we first de-
veloped to tap intrinsic versus extrinsic orientation in the classroom (Har-
ter, 1981), pitting these orientations against one another in the following
manner. We first identified three components: (a) preference for challenge
versus preference for easy work assigned, (b) curiosity or intrinsic interest
versus getting grades/pleasing the teacher, and (c) independent mastery
versus dependence on the teacher. Thus, we were interested in (a) does the
child like hard, challenging work or does the child like the easier assign-
ments and school subjects, (b) does the child work to satisy his/her own
interest and curiosity or does the child do schoolwork in order to merely
satisfy the teacher and obtain the desired grades, and (c) does the child
prefer to figure out problems or difficult assignments on his/her own or
does the child rely on the teacher for help and guidance, particularly when
it comes to figuring out challenging problems or assignments.
The particular question format we developed for this instrument direct-
ly pits these orientations against one another, as alternative choices. Thus,
children are presented with a two-part statement that reads: If a school
subject is hard to understand, some kids want the teacher to explain it to
them BUT Other kids would like to understnd it for themselves. The re-
spondent is first asked to decide which of the two alternative children de-
Teacher and classmate influences 13
scribed in each part of the item are most like him/her, and then asked to
indicate whether that statement is "Really true for me" or just "Sort of
true for me." Items are scored on a four-point scale, where 4 is the most
intrinsic and 1 is the most extrinsic. It should be noted that the question
content is not specific to any particular school subject matter or context,
since it was our intent to assess a child's general orientation to classroom
learning.
With regard to developmental differences, our own work (Harter,
1981) has demonstrated a systematic, grade-related shift from a predomi-
nantly intrinsic motivational orientation in third grade to a more extrinsic
motivational orientation by ninth grade. Scores in third grade (M= 3.10)
reveal that children are intrinsically motivated. However, scores systemat-
ically decline over the grade school years, such that in junior high school,
scores hover around 2.25, revealing an extrinsic orientation. The biggest
drop is between sixth grade (M= 2.60) and 7th grade (M= 2.30).
Other cross-sectional developmental data document a similar pattern
of decreasing intrinsic interest in learning as a function of grade level,
particularly with the shift to junior high school (Brush, 1980; Eccles &
Midgley, 1988, 1990; Eccles et al., 1984; Harter, 1981; Simmons & Blyth,
1987; Simmons, Blyth, & Carleton-Ford, 1982; Simmons, Rosenberg, &
Rosenberg, 1973). Gottfried (1981) has documented such a decline in in-
trinsic motivation at seventh grade for specific school subjects, namely
reading, science, math, and social studies.
Several psychological analyses of the causes of such shifts have now
emerged. In our own work, we initially speculated that students face the
demands of a school culture that increasingly reinforces ans external mo-
tivational orientation, especially through grading practices (Harter, 1981).
Thus, as the educational focus gradually shifts to the products of one's
learning, evaluated through grades, children adapt to this reward system,
and their interest in the learning process itself declines.
Eccles and her colleagues (Eccles & Midgley, 1988, 1990; Eccles et
al., 1984) have provided a much more complete analysis of the possible
reasons why children's attitudes toward school learning and achievement
become increasingly negative as they progress through the school system,
identifying specific changes in the school environment as critical. Their
earlier reviews suggested that with the transition to junior high school, the
school environment becomes more impersonal, more formal, more evalu-
ative, and more competitive than in the elementary grades. It is primarily
the teachers who communicate these changing values and standards.
There is also increasing emphasis on social comparison as students come
to be graded in terms of their relative performance on standardized tests

and assignments and as information about individuals' performance levels
becomes more public. Moreover, teachers are perceived as having increas-
ing control over evaluative outcomes. Nicholls (1979) specifically high-
lights the implications of such changes for the self, suggesting that they
lead students to focus on the assessment of their ability, rather than on the
learning task itself. He argues that this change in focus, in turn, has a neg-
ative impact on children's motivation to learn.
More recently, Eccles and her colleagues (Eccles & Midgley, 1988,
1990) have elaborated on their analysis in the form of a model of person-
environment fit. They suggest that the lack of fit between the junior high
school environment and the needs of young adolescents contributes to the
documented shift toward more negative self-evaluations and attitudes to-
ward school learning. They note that teachers become more controlling,
just at the point that adolescents are seeking more autonomy. In addition,
the teacher-student relationship is becoming more impersonal at the same
time that students, in their bid for autonomy, increasingly need the person-
al support of adults other than their parents. Finally, various forms of so-
cial comparison are increasingly promoted, just as adolescents are enter-
ing a period of heightened self-consciousness.
Although these analyses provide a general framework for understand-
ing grade-related changes in motivational orientation, they do not speak to
(a) the longitudianl study of actual change over time within the same indi-
viduals, to (b) possible individual differences in students' reactions to edu-
cational transitions, or to (c) students' perceptions of the postulated
changes in the academic environment. In our most recent work, we have
begun to address each of these concerns.
Patterns of change upon entrance to junior high school

Employing a seven-month longitudinal design, we first examined the mo-
tivational orientation of 6th grade elementary school children in the spring
of the academic year, before their transition to junior high school. We then
retested them in December of the following school year, several months
after the shift to the new school environment. Based on preliminary find-
ings, we questioned whether all students would react to such a transition
with declines in intrinsic motivation, as the literature had implied. Rather,
we suspected that individual children would react differently to the envi-
ronmental changes accompanying school transitions. More specifically,
we anticipated that three groups of students would emerge: those whose
intrinsic motivation would decrease, those whose intrinsic motivation

would actually increase, and those who would report no change in motiva-
tional orientation. This is precisely the pattern that the data revealed (Har-
ter, 1992; Harter, Whitesell, & Kowalski, 1992).
With regard to the factors that may cause or produce stability or insta-
bility in motivational orientation over time, we have focused on how edu-
cational transitions initially impact perceived scholastic competence,
which in turn influences motivation. Our studies have consistently re-
vealed that motivational orientation and perceived scholastic competence
are very highly related (Harter, 1981a; 1981b; Harter & Cornell, 1984;
Harter & Jackson, 1992). Thus, students who evaluate their competence
positively will typically report that they are intrinsically motivated, where-
as those with negative perceptions of their competence will invariably re-
port that they are extrinsically motivated.
With regard to factors that impact students' perceptions of their compe-
tence as a function of the transition to junior high school, our model identi-
fies the influences of both teachers and peers. Specifically, we anticipated
that with the transition to junior high school, particular teacher and class-
mate factors would serve to cause scholastic competence to become more
salient as well as susceptible to change. With regard to the impact of teach-
ers, our analysis draws upon the observations of Eccles and her colleagues,
cited earlier (Eccles et al., 1984; Eccles & Midgley, 1988, 1990). Thus, we
anticipated that hypothesized grade-related change in educational prac-
tices, namely, teachers' greater emphasis on grades, their greater focus on
competition and on control, which leads to heightened external evaluation
of performance, coupled with teachers' decreasing personal interest in stu-
dents, should lead students to reevaluate their sense of competence.
Classmate influences should also contribute to such reevaluation. Typi-
cally, there are several elementary feeder schools from which junior high
school students are drawn. A significant portion of the seventh grade ju-
nior high school class, therefore, will be students whom one did not previ-
ously know. Thus, the social reference group is widely expanded, requir-
ing that students reassess their competence in relationship to members of
this new social comparison group. This increasing emphasis on social
comparison within the junior high school environment also serves to fos-
ter a reevaluation of one's scholastic competence, relative to the abilities
of the new cohort of classmates. These comparisons can have devasting
psychological effects for a large number of students who conclude that
they are relatively incompetent, compared to those at the top. This general
model is presented in Figure 2.1.
TEACHER INFLUENCES:
- Greater emphasis on grades

- More academic competition
- More teacher control
- Less personal interest
RE-EVALUATION OF PARALLEL CHANGES IN

SCHOLASTIC COMPETENCE MOTIVATIONAL ORIENTATION
CLASSMATE INFLUENCES:
- New social reference group

- More social comparison
Figure 2.1. Model of teacher and classmate influences on perceived scholastic competence and motivational orientation.
In our longitudinal study we anticipated, therefore, that a certain per-

centage of students facing this school transition would alter their percep-
tions of competence, some feeling more competent in the new environ-
ment and some feeling less competent. We also expected that there would
be a subgroup of students who would maintain stable perceptions of com-
petence. Our findings supported these predictions in that approximately
50 percent of the students reported either substantial increases or decreas-
es in their perceptions of competence, whereas the remaining 50 percent
maintained relatively stable estimates of their academic abilities.
The hypothesis that change in perceptions of competence should pre-
dict parallel changes in motivational orientation, was also supported by
our findings, as can be seen in Figure 2.2. Increases in perceived compe-
tence were associated with increases in intrinsic motivation, decreases in
perceived competence predicted declines in intrinsic motivation, and stu-
dents reporting no change in perceived competence manifested negligible
changes in motivational orientation. One could certainly argue that the
causal factors on our model, namely teacher and classmate influences,
also have a direct impact on motivational orientation, undermining intrin-
sic motivation for many, but by no means all, students. However, that the
mediational role assigned to perceived competence would seem to best
explain the individual differences in motivational orientation that we have
obtained.
Students awareness of changes in the educational environment

Our interpretation of transition-related changes in children's academic
self-concepts and motivational orentation rests heavily on the assumption
that changing educational practices instituted by teachers caused the eval-
uation of one's ability to become more salient, particularly in the face of a
new social reference group to whom one's performance can be compared.
However, do students actually experience the types of alterations in the
academic environment that Eccles and her colleagues have identified?
Since students are the direct recipients of the educational practices that
have been interpreted as changing, it seemed essential to examine stu-
dents' perceptions directly.
To address this issue, we conducted a follow-up study in which we
asked middle school students to compare their school environment in the
current year to that of the previous year (Harter, Whitesell, & Kowalski,
1992). Students rated a number of dimensions, including importance of
grades, level of competition, teachers' emphasis on the products of learn-
Perceived >
Competence /
HI Increased /
H 3.00-
\ /
CL
O / Remained
O . the
CO \ Same
<T 2.50
\
o
o k
CO
\
VEC
\
UU 2.00 - Perceived ^
O Competence \
LU Decreased \
Q.
1.50 - 1
Sixth Seventh
Grade Grade
Timel Time 2
3.50
Perceived
Competence
CO Increased
E
3.00
2.50 •
CO Perceived
Z 2.00 Competence
s
a: Decreased
1.50
Sixth Seventh
Grade Grade
Time 1 Time 2
Figure 2.2. Relationship between changes in perceived competence and

changes in motivational orientation.
ing (i.e., knowing the correct answer), teacher control, preoccupation with
ability, and social comparison. They provided two sets of ratings, one for
their current educational environment, and one for the previous school
year.
The findings revealed that the vast majority of middle school adoles-
cents reported the types of changes observed by Eccles and her col-
leagues. In comparing educational practices in their current school envi-
ronment to those in the previous school year, these students perceived a
definite increase in competition, focus on grades, external evaluation,
teacher control, and concern with their academic ability. These subscales,
which formed one factor, were rated 3.23 (on a four-point scale) for stu-
dents' current school year compared with 2.83 for the previous school
year. Moreover, the absolute values of these scores are also noteworthy,
since they document students' perceptions of considerable emphasis on
external evaluation.
Less dramatic, although significant, increases in social comparison
were reported (2.88 for the current school year compared with 2.60 for the
previous school year). We suspect, however, that social desirability re-
sponse tendencies restrain students from admitting that they use social
comparison information. Ruble (1983) has reported that while social com-
parison increases as students move up the academic ladder, students' will-
ingness to admit that they are engaging in it decreases. In previous stud-
ies, we have also found that students acknowledge only modest use of
social comparison information. However, when we have asked young ado-
lescents to give specific examples of classmates at six different levels of
scholastic competence, they have no trouble identifying the academic lev-
el of every student in their class! Thus, if is clear that they are well aware
of differences in the scholastic ability of their peers, and where they stand
by comparison.
In general, therefore, it would appear that students recognize that the
educational climate shifts as they move up the acadimic ladder. They re-
port their awareness of a constellation of changing teacher behaviors that
undermine perceptions of competence and intrisic motivation as well and
increased use of social comparison. These perceived changes would ap-
pear to provide sufficient impetus for students to reexamine, and possibly
alter, their perceptions of competence, thus producing the pattern of indi-
vidual differences to school transitions that were reported above.
We have, most recently, refined our measure of students' perceptions
of the educational climate that is fostered by teachers and classmates, in
order to further investigate these issues. We were particularly interested in
students' perceptions of why students lose their intrinsic motivation. Thus,

we asked them to respond to a number of reasons why they felt that stu-
dents get turned off to schoolwork, why they may not be interested in
learning, why they wouldn't be curious, and why they might not want to
figure things out on their own. This most recent instrument contains seven
subscales, and places more direct emphasis on the role of teachers as the
conveyers of educational values and standards. Four of the subscales de-
fine one factor that concerns dimensions of external evaluation. These
were (a) Grades ("The teachers put too much emphasis on grades";
"Teachers seem more concerned about the grades students get than about
what they are learning"); (b) Competition ("Teachers expect students to
compete too much with their classmates"; "Students have to compete with
each other to get good grades"); (c) Control/choice ("The teachers don't
give students enough say or choice in the schoolwork they do"; "Teachers
don't let students decide what work they want to do"); (d) Lack of person-
al interest ("Teachers don't really care that much about the students and
their feelings"; "Teachers don't seem that personally interested in students
as people").
A separate subscale tapped social comparison ("The teachers let stu-
dents know how well everyone in the class is doing"; "Teachers make stu-
dents' grades too public by letting others know how they have done on
tests and assignments"). In addition, we added two new subscales. One
tapped the extent to which students were turned off to schoolwork because
it was deemed boring or irrelevant ("Most of the schoolwork is pretty bor-
ing"; "Schoolwork just isn't that important or relevant to students' lives").
Another directly tapped the perception that feeling stupid was a motive for
becoming turned off to schoolwork ("The students are not doing that well
and it makes them feel stupid"; "Students feel like they are dumb because
they don't do well at their schoolwork").
We examined these components in a large middle school population
(6th, 7th, 8th grades) drawing from lower-middle and middle-class fami-
lies. Figure 2.3 reveals systematic grade differences for all four compo-
nents, where the differences between sixth and seventh grades are some-
what bigger than between seventh and eighth grades. With increasing
grade level, students perceive teachers to be more evaluative. In addition,
students report increasing emphasis on classmate social comparison. In-
terestingly, the higher the grade level, the more students feel that school-
work is boring and irrelevant. Finally, the higher the grade level, the more
students endorsed "feeling stupid" as a motive causing one to be turned
off to schoolwork. Of further interest was the finding that the more stu-
3.50 T
Feel Stupid
Irrelevant
3.00 -
Soc. Comparison
Evaluation
2.50
2.00
Sixth Seventh Eighth

GRADE LEVEL
Figure 2.3. Relationship between changes in perceived competence and changes in motivational orientation.
dents felt that teachers emphasized the components of external evaluation,

the more they reported feeling stupid as a motive undermining intrinsic
interest (r = .48). Thus, the pattern supports the contention that students
are aware of the dimensions that have been hypothesized to undermine
perceptions of both scholastic competence and intrinsic motivation, and
feel thay they increase in intensity across the middle school years.
In this study we also asked teachers to rate these same items, ratings
which we could compare to students' perceptions. Not surprisingly, per-
haps, teachers (M= 2.07) did not feel that they were emphasizing the ex-
ternal evaluation components nearly as much as did students (M = 2.69).
Nor did teachers (M= 2.22) feel that they were making social comparison
as salient as did students (M= 2.81). Teachers (M= 2.91) did agree with
the perceptions that students feel that schoolwork is boring and irrelevant
(M = 2.87). In so doing, teachers would appear to be placing the onus for
lack of interest in learning on students who fail to see the importance of
schoolwork. Finally, teachers ( M = 2.86) and students (M = 2.90) agreed
that if students felt stupid, it would undermine their intrinsic interest in
schoolwork. However, there was no evidence that teachers felt that their
educational practices might contribute to students' experience of stupidity.
In this same study, we were interested in whether there was a relation-
ship between students' perceptions of their educational environment and
their own self-reported motivational orientation. We hypothesized that
students low in intrinsic motivation (and high in extrinsic motivation)
would be more likely to report negative environmental influences than
would students high in intrinsic motivation (and low in extrinsic motiva-
tion). That is, those low in intrinsic motivation (and high in extrinsic moti-
vation) would be more likely to report that teachers' focus on external
evaluation (for the combination of emphasis on grades and competition,
lack of student choice and personal interest in students), that social com-
parison is more salient, that schoolwork is irrelevant or boring, and that
feeling stupid will undermine one's interest in school learning. Our pre-
dictions were based on the assumption that students who perceived their
environment to stifle intrinsic interest in learning would be less likely to
display such interest.
As Figure 2.4 reveals, this is precisely the pattern we obtained. All
comparisons revealed marked differences that were highly significant (p <
.001). Moreover, the general finding that lowered perceptions of scholas-
tic competence are associated with a more extrinsic orientation (M= 2.63)
was replicated. Intrinsically-oriented students reported considerably high-
er levels of competence (M= 3.38). With regard to the focus of this chap-
3.50 T
H Low Intrinsic/High Extrinsic
• High Intrinsic/Low Extrinsic
3.00
2.50
2.00
Evaluation Soc. Irrelevant Feel Stupid

Comparison
Figure 2.4. Perceptions of the social/educational environment as a function of motivational orientation.
ter, noteworthy were the effects obtained for the teacher evaluative com-
ponents and for social comparison with classmates. These findings are
open to a complex set of interpretations, since students of both types, high
intrinsic and low intrinsic motivational orientations, can be found within
the same classroom. Why should different students perceive the environ-
ment so differently? Is it the case that student perceptions of their
social/educational environment is colored by their own particular motiva-
tional orientation? Thus, are extrinsically motivated students more likely
to report that teacher attitudes and practices undermine their interest in
learning? Or might it be that teachers respond differently to students who
display different motivational styles and accompanying levels of scholas-
tic competence? Do teachers react more positively to competent students
who are intrinsically motivated? Alternatively, if educational practices
heighten the salience of one's competence, might those students who feel
they fall short, in comparison to classmates, adopt an exaggerated view of
the detrimental features of their environment? Clearly we need further re-
search that more objectively documents the actual behaviors of teachers
within the classroom setting, and which can speak to the directionality of
these effects. However, we feel that our findings address the issue of how
students make meaning out of their own experiences, and as such they of-
fer valuable insights into the perceived social and educational milieu in
which students must function.
Self-esteem
Another constuct that can be profoundly influenced by factors within the
classroom setting is self-esteem. To understand these processes we must
first turn to those historical scholars who identified the critical determi-
nants of individuals' global evaluation of their worth as a person. William
James (1892) put forth a compelling formulation on the causes of self-
esteem. For James, this self-evaluative process involved an examination of
the ratio of one's perceived successes to one's pretensions to be successful.
Thus, if perceived successes were equal to or greater than one's aspira-
tions for success, high self-esteem would result. Conversely, if one were
unsuccessful in domains where one had pretensions to be successful, low
self-esteem would ensue.
In contrast to James, symbolic interactionists such as Cooley (1902)
and Mead (1925; 1934) placed heavy emphasis on how one's social inter-
actions with others profoundly shaped the self. Thus, for these scholars,
the self is considered to be primarily a social construction, crafted through
linguistic exchanges (symbolic interactions) with others. Cooley's formu-

lation was perhaps the most metaphorical given his postulation of the
"looking glass self." For Cooley, significant others constitute a social mir-
ror into which the individual gazes in order to detect their opinions toward
the self. These opinions, in turn, are incorporated into one's sense of self.
Cooley contended, therefore, that what becomes the self is what we imag-
ine that others think of us, including our appearance, motives, deeds, char-
acter, etc. One comes to own these reflected appraisals.
In Mead, we find an elaboration of the themes identified by Cooley,
with an even greater insistence on the role of social interaction, particular-
ly through the use of language. For Mead, "We appear as selves in our
conduct insofar as we ourselves take the attitude that others take toward
us. We take the role of what may be called the "generalized" other. And in
doing this we appear as social objects, as selves" (p. 270).
Our own research has found strong support for both sources of self-
esteem, namely perceptions of success in areas where one has aspirations
for success (James) and the internalization of the approval (or disap-
proval) of significant others (Cooley and Mead). Across numerous studies
(Harter, 1986, 1990, 1993) we have demonstrated that perceptions of
competence or adequacy in domains deemed important as well as the sup-
port of significant others are each strongly predictive of global self-
esteem. The domains we identified for older children and adolescents in-
clude scholastic competence, athletic competence, peer likability, behav-
ioral conduct, and physical appearance. The particular sources of support
include parents, teachers, classmates, and close friends. Most relevant to
this chapter is the domain of scholastic competence and the support from
teachers and classmates.
The contribution of perceived scholastic competence to self-esteem

With regard to the impact of perceived scholastic competence, across
samples of older children and adolescents, correlations with self-esteem
range from .46 to .64. When we focus specifically on those students for
whom scholastic competence is judged important, the magnitude of these
correlations (rs from .53 to .68) increases, consistent with James' formu-
lation. The reason that these latter values are not dramatically higher is
that the vast majority of children report that doing well academically is
important to them. That is, there is considerable overlap (approximately
85%) between the group containing all subjects and the group who report
that scholastic competent is important. The more critical test of James'
theory involves an examination of the correlation between perceived com-

petence and self-esteem among those who indicate that scholastic compe-
tence is not important. Among this minority (approximately 15%) the cor-
relation falls to between .23 and .28, across samples.
How does the impact of perceived scholastic competence compare to
other domains we have included? We consistently find that perceived
physical appearnace correlates most highly with self-esteem, across the
lifespan (rs from .65 to .82). Among older children and adolescents,
scholastic competence and peer acceptance are the next most predictive
domains. The correlations of behavioral conduct and athletic competence
are somewhat lower, although they also contribute significantly.
The fact that the vast majority of students report that scholastic compe-
tence is important, and that among these students, correlations with self-
esteem are not only significant but substantial, as reported above, under-
scores the impact of one's perceived academic success or failure on
student's psychological adjustment. An inspection of individual students
with low self-esteem reveals that for many there is a dramatic discrepancy
between the importance that they attach to academic success and their
perceptions of scholastic incompetence. This discrepancy, as James ar-
gued, seriously erodes their global feelings of self-worth. James' formula-
tion also implies two strategies for enhancing one's self-esteem. One can
reduce such a discrepancy by either increasing one's competence or by
discounting the importance of the domain.
However, elevating one's actual level of competence may meet with
roadblocks, to the extent that there may be natural limits on one's ability.
Even in the face of effort and some demonstrable, absolute level of im-
provement, social comparison standards may preclude dramatically en-
hanced perceptions of one's relative competence. Moreover, it is very dif-
ficult for students to discount the important of scholastic competence,
given that academic success is touted as critical by most parents, teachers,
and by society, in general. Thus, despite the hypothetical pathways to self-
esteem enhancement, the realities place limitations on the ability to in-
crease one's sense of worth, particularly if the discrepancy between the
importance of schoolwork and perceived scholastic competence is a major
contributor to low self-esteem.
The contribution of teacher and classmate support to self-esteem

In addressing the symbolic interactionists' contention that approval from
significant others is incorporated into one's sense of self-esteem, we have
examined the correlations between perceived approval from parents,

teachers, classmates, and close friends. Beginning in late childhood and
continuing into adolescence, classmate support correlates most highly
with self-esteem (rs range from .50 to .60). Parents support follows close-
ly (rs range from .49 to .56). Teacher approval also correlates significantly
(rs range from .40 to .45) reflecting the importance of teacher support to
child and adolescent self-esteem. Close friend support correlates least
highly of the four sources (rs range from .35 to .40).
Clearly, peer support in the form of approval from classmates within
the school contexts (in contrast to approval from close friends) is critical
to self-esteem. It is of interest to note that at every developmental level we
have investigated, namely middle to late childhood, adolescence, the col-
lege years, and early to middle age adulthood, we have consistently and re-
peatedly found that approval from peers in the more "public domain"
(e.g., classmates, peers in organizations, work settings, etc.) is far more
predictive of self-esteem than is approval from one's close friends. We in-
terpret this finding to suggest that support from others in the more public
domain may better represent acceptance from the "generalized other"
(Mead, 1934), approval that may be perceived as more "objective" or from
more credible sources than the support from one's close friends. This is
not to negate the importance of close friend support, which would appear
to be critical as a source of acceptance, feedback, and clarification of val-
ues vis-a-vis the outside world. Such close friend support would seem to
function as a secure psychological base from which one can reemerge to
meet the challenges of the generalized other, whose acceptance appears
critical to maintaining high self-esteem.
Among older children and adolescents, it is one's classmates who pri-
marily represent this more public, "generalized other." Thus, the peer en-
vironment within the school context looms large as a critical determinant
of one's sense of worth as a person. Moreover, it is not an environment
from which one can escape, if classmate support is not forthcoming. On a
daily basis, low self-esteem children must confront the disapproval, rejec-
tion, or neglect of the very peers whose support is critical to their sense of
personal worth.
Findings from our laboratory reveal reasons why it may be difficult for
such children to obtain classmate approval. We have asked subjects to rate
the importance that they feel their classroom peers attach to the domains
we have examined. They report that classmates place the most importance
on physical appearance, likeability, and athletic competence (whereas they
feel that parents value scholastic competence and behavioral conduct
more highly). Thus, in order to obtain classmate approval, one must be

good looking, likeable, and athletically talented. Our more comprehensive
causal modeling efforts further confirm this conclusion in that perceived
adequacy in these three domains is highly predictive of classmate support
(Harter, Whitesell, & Marold, 1991). However, as noted in the discussion
of perceived scholastic competence, there may be natural limitations to
one's ability to improve in these areas, preventing them from garnering the
classmate approval that is so critical to self-esteem. (See other chapters in
this volume by Erdley, Juvonen, Kindermann et al., Kupersmidt et al., and
Ladd & Birch, which offer additional models for children's lack of peer
support).
Within the school context, it is classmate support that is most critical to
self-esteem; however, the impact of teacher support is not trivial. The cor-
relations between teacher approval and self-esteem may actually underesti-
mate the influence of a given teacher in a child's life, since our questions
asked them about support from teachers in general. Anecdotal evidence
makes it clear that among middle school and junior high school students,
who have different teachers for different school subjects, some teachers
may be perceived as far more supportive than others. Our procedure did not
allow us to make this discrimination, however. Thus, in the future it would
be instructive to have separate ratings on different teachers in order to de-
termine whether some have a greater influence on self-esteem than others.
Evidence from our laboratory further reveals that teacher support may
serve to compensate for low parent support. Thus, students who reported
both low parent and low teacher support had self-esteem scores of 2.57.
However, students with low parent support who reported high teacher sup-
port had self-esteem scores of 2.80, a significant difference. The additive
nature of parent and teacher support could be observed at moderate and
high levels of parent support, where higher teacher support resulted in
higher levels of self-esteem.
The directionality of the link between classmate approval

and self-esteem
Although the correlation between approval from various sources and self-
esteem would seem to support the symbolic interactionist position in
which the opinions of others are presumed to impact one's self-evaluation,
one can also raise the issue of whether one's self-evaluation impacts one's
perceptions of the opinions of others. Felson (1993) addresses this very is-
sue. He observes that it may well be that, in contrast to the looking glass
formulation in which the opinions of others are internalized and therefore

dictate one's level of self-esteem, that one's level of self-esteem may be
driving what one perceives to be the opinions of others. He reports a series
of studies revealing that children who like themselves assume that their
parents' and peers' reactions to them are favorable, leading to an alterna-
tive interpretation which he terms the "false consensus effect." That is,
one falsely, perhaps, assumes that others share your own view of the self.
Although the label "false" seems a bit pejorative, it is highly likely that
both effects, internalization of others' opinions as well as assumptions that
if one likes the self, others in turn will manifest their approval, are opera-
tive.
We became particularly interested in this issue with regard to the "gen-
eralized other," operationalized as classmates in our studies. We have dis-
covered that there are individual differences in the directionality of the
link between classmate approval and self-esteem, differences which have
a predictable pattern of correlates (Harter, Stocker, & Robinson, in press).
We first became intrigued with the issue of whether adolescents con-
sciously endorse a looking glass self metatheory, meaning that they ac-
knowledge that they need to have other people evaluate them positively in
order to like themselves as a person (our definition of self-esteem). The
competing metatheory is that if one likes oneself as a person, then others
will necessarily like or support the self, as well. Thus, with regard to the
directionality of effects, in the first case, approval precedes self-esteem
whereas in the second, self-esteem precedes approval.
We initially demonstrated that there are adolescents who endorse each
of these metatheories about the directionality of the causal links between
approval and self-esteem. Moreover, a third group emerged, namely ado-
lescents who reported that there was no relationship between the approval
of others and how much one liked oneself as a person. Important to these
distinctions were adolescents' open-ended verbal descriptions, which con-
verged with their directionality choice. Looking-glass self subjects gener-
ated explanations such as: "If other kids approve of me and say good
things about me, then I look at myself and think I'm not so bad, and I start
liking myself"; "By having other people approve of you it gives you self-
confidence and so you like yourself"; "If other people don't like me as a
person, then I wonder if I am a good person, I care about what people say
about me"; "If nobody liked you, you probably wouldn't like yourself very
much." It is of interest that this group appears to couch their descriptions
in the language of the "generalized other," making reference to people,
kids, and to others in general, rather than particular people.
Those reporting that their self-esteem precedes approval from others

provided such descriptions as: "If I like myself as a person, then I know
that the other kids will feel pretty much the same"; "You have to appreci-
ate yourself first, as a person, if you wait for other people to make you feel
good, you could be waiting a long time!"; "The way I figure it, if you can't
like the person you are first, then how do you expect other people to like
you?"; "If you like yourself, and feel good about yourself, other kids will
like you more."
The third group denied that there was any link between approval from
others and self-esteem. Sample descriptions were: "If someone doesn't
like me, they can't make me feel bad if I'm confident enough about my-
self, what they think doesn't matter"; "I really don't care if someone does-
n't like me because its their problem"; "It doesn't matter what other peo-
ple think of me, it doesn't have anything to do with how much I like
myself."
Having established the meaningfulness of these three groups of adoles-
cents, each of whom had very different metatheories about the link be-
tween approval and self-esteem, we next sought evidence for a predicted
pattern of correlates. We specifically hypothesized that there would be
certain liabilities associated with the endorsement of a looking-glass per-
spective, at this particular developmental level. The pattern of findings
supported our predictions. First, those endorsing the Approval to Self-es-
teem directionality (the looking-glass self subjects) reported significantly
greater preoccupation with peer approval than the Self-esteem to Approval
group, with the No Connection group reporting the least preoccupation.
The findings also revealed that the looking-glass self subjects reported
greater fluctuations in perceived peer approval and self-esteem, compared
to either of the other two groups. While adolescence, as a general develop-
mental period, tends to usher in volatility in the self-concept (Harter,
1986; Harter & Monsour, 1992; Rosenberg, 1986) those basing their self-
esteem on peer approval would appear to be particularly vulnerable to
changes in self-esteem. Given that they are basing their self-esteem on
what they perceive to be fluctuating peer approval, it is understandable
why their self-esteem should fluctuate, in tandem.
Of further interest was the finding that looking-glass self adolescents
also reported a lower level of peer approval than those in the Self-esteem
to Approval group. Given this pattern, we also hypothesized that the level
of self-esteem should be lower for the looking-glass self subjects, given
that by definition, they are basing it on peer approval. Finally, the liabili-
ties of a looking-glass self orientation extend beyond mere perceptions, in
that teachers observed behavioral ramifications in the classroom. Teach-

ers' ratings revealed that the looking-glass self students were more likely
to be socially distracted from their scholastic activities, which may be due
to their preoccupation with peer approval. (The No Connection group
were the least distracted.)
The pattern reveals that there are clear liabilities associated with the
endorsement of a looking-glass self model in adolescence. It is important
to point out that developmentally, a looking-glass self model represents an
important mechanism through which opinions of others come to impact
the self. However, the healthiest developmental course would appear to be
one in which the standards and opinions of others are eventually internal-
ized, such that truly se/^evaluations become the standards that guide be-
havior. When or if one reaches this stage, one endorses a directionality
orientation in which how one evaluates the self will impact others' opin-
ions of the self.
What is it that might prevent individuals who endorse a looking-glass
self model in adolescents from pursuing the path to internalization? At
this time, we can only speculate. The developmental history of such indi-
viduals may have been characterized by more parental disapproval, incon-
sistent or fluctuating approval, or support that was conditional upon meet-
ing the demands of others. Such a history could lead to preoccupation and
concern over support, perceptions of lower and fluctuating support, which
in turn would lead to lower and fluctuating self-esteem. There would also
appear to be some parallels between the looking-glass self group and the
"preoccupied" or anxious ambivalent attachment style, as well as between
the Self-esteem Precedes Approval group and the securely attached style.
Moreover, many of the No Connection style subjects who indicated that
they did not care about the opinions of others suggests a defensive posture
reminiscent of the "dismissing avoidant" attachment style (Bartholomew,
1990). Thus, future research may do well to focus on those child rearing
antecedents identified by attachment theorists, given their view that early
relationships with caregivers provide working models of self and others
that children carry into relationships with peers.
Suppression of voice
The third construct to be examined is students' ability to express their
opinions. Our interest in this topic emerged in the course of our research
on the emergence of "false self behavior" during adolescence (Harter,
Marold, Whitesell, & Cobbs, in press). False self behavior involves acting
in ways that do not reflect one's true self as a person, the "real me." In
their open-ended descriptions, adolescents frequently describe false self
behavior as "not saying what you think," "expressing things you don't re-
ally believe or feel," "not stating your true opinion," "saying what you
think other people want to hear." These observations converge with what
Gilligan and her colleagues (Gilligan, 1982; Gilligan, Lyons, & Hanmer,
1989) have referred to as "loss of voice," namely the suppression of one's
thoughts and opinions. Gilligan finds loss of voice to be particularly prob-
lematic fox females, beginning in adolescence.
Gilligan observes that before adolescence, girls seem to be clear about
what they know and most are able to express their opinions forcefully.
However, with the onset of adolescence, many cover over what they knew
as children, suppressing their voice, diminishing themselves, and hiding
their feelings in a "cartography of lies." For Gilligan, there are several
compelling motives leading females, beginning in adolescence, to silence
their voices. During this developmental transition, adolescent girls begin
to identify with the role of "woman" in the culture. They quickly perceive
that the desirable stereotype of the "good woman" involves being nice,
polite, pleasing to others, and quiet if not shy. In what is still largely a pa-
triarchal society, they observe that women's opinions are not sought after,
not valued, not supported. Therefore, their own voice begins to go under-
ground. Moreover, many girls observe these behaviors in their own moth-
ers, who serve as role models for how women in this culture should act.
Thus, they identify with their mothers, whom they observe to be suppress-
ing their own opinions; they accept their role as women, and eventually
come not only to not speak their minds but to not know their own minds.
Thus, they no longer even have an opinion.
Gilligan also describes another constellation of proximal motives that
derive from the relational impasse in which adolescent girls find them-
selves. Given the importance of connectedness to others for females, in
particular, behaviors that threaten relationships are to be avoided at all
costs. Thus, beginning in adolescence, females compromise their authen-
ticity, they take themselves out of "true relationship," in order to preserve
connectedness in some lesser form. If they were to speak their mind, ex-
press their true voice, it may well cause tension in the relationship, make
the other person angry, hurt others' feelings, and, at worst, lead others to
abandon them altogether. Thus, they opt not to take the risk of rejection or
relationship dissolution.
Gilligan's analysis is quite provocative, and her observations, including
interviews and dialogues with adolescent girls, clearly reveal that for
many female adolescents, suppression of voice becomes the only path
through which they can preserve relationships. However, Gilligan's efforts

have not resulted in any systematic analysis of the prevalence of loss of
voice among females, nor has she chosen to address these issues in ado-
lescent males. To what extent are their individual differences in the ex-
pression of voice, and what factors might be predictive of such differ-
ences, in males as well as females?
We have recently begun to address these issues in our own research. In
our initial study, we included young adolescent females and males, ages
12 through 15, in grades 6,7, and 8 of a middle school (Harter, Waters, &
Gonzales, 1994). We were particularly interested in level of voice in three
different relational contexts, namely, with close friends, with peers (peo-
ple of the same age who were not close friends), and with teachers in the
classroom. A sample item tapping level of voice with peers was as fol-
lows: "Some kids usually don't say what's on their mind or what they're
thinking to people their age BUT Other kids do say what's on their mind or
what they're thinking to people their age." Subjects first select the kind of
kids that they are most like, and then indicate whether that choice is "Re-
ally True for Me" or "Sort of True for Me." Items are scored on a four
point scale where 1 equals the lowest level of voice and 4 the highest. (The
content of the items was identical for close friends and teachers.)
In order to begin to explore the potential causes of individual differences
in level of voice, we speculated that endorsement of gender role stereotypes
in the form of femininity, masculinity, and androgyny may well be an im-
portant predictor. We hypothesized that for females, in particular, individ-
ual differences in level of voice would be related to their sex-role orienta-
tion. Specifically, those endorsing a feminine orientation (and eschewing
the masculine role model) should report greater loss of voice than those en-
dorsing either an androgenous (high femininity, high masculinity) or a mas-
culine-only orientation. We reasoned that those with a feminine orientation
would be most likely to adopt the relational values of the "good female" in
our society, which in turn would dictate suppression of voice.
Feminine sex-role items adapted from Boldizar (1991) included "I care
about what happens to others," It makes me feel bad when someone else is
feeling bad," "I am a gentle person," "I'm good at understanding other
people's problems," "I like small babies and children a lot," "I am a kind
and caring person." Masculine items included "When I play games, I like
to win," "I can control the other kids in my class," "I make a strong im-
pression on most people I meet," "I am sure of my abilities," "I am a
leader among my friends."
With regard to level of voice across the three different relationships,
the findings revealed that overall, young adolescents are most outspoken
with their close friends (M= 3.11), followed by peers (M= 2.75), and then
teachers (2.63). As Gilligan has observed, it is within one's relationship
with closest friends that voice should most likely be expressed, consistent
with our findings. Levels of voice were virtually identical for boys and
girls in their relationships with close friends and peers. Thus, in these con-
texts there is no evidence, as Gilligan implies, that loss of voice is a par-
ticular problem for females. However, with regard to voice with teachers,
the adolescent females did report significantly less voice (M= 2.49) than
did males (2.77). Just why female students, in particular, are reluctant to
express their opinions to teachers is an interesting topic to which we shall
return.
Of greatest interest were the findings related to sex-role orientation,
particular with regard to voice with peers. As hypothesized, those females
who were high on femininity and low on masculinity reported significant-
ly lower levels of voice (M= 2.49) than those who were androgenous (M =
2.86). The small number of females reporting only a masculine orientation
reported even higher levels of voice (M= 3.06). Thus, it definitely appears
that gender role orientation is a powerful factor determining individual
differences in voice among female adolescents. For adolescent males, the
pattern was somewhat comparable in that the small number of boys who
endorsed a feminine orientation reported lower levels of voice (M= 2.73)
than those endorsing either the androgenous (M= 2.93) or the masculine
(M= 2.97) orientation.
The findings suggest, therefore, that gender role orientation, rather
than gender, per se, appears to be a major factor contributing to level of
voice among young adolescent males and females. Those of either gender
who adopt feminine sex-role stereotyped behaviors are at greater risk for
lower levels of voice. Whether those who suppress their voice in adoles-
cence have actually lost their voice is an interesting question for further
study. That is, at what point in development might such a relationship
emerge? Might it be that we would find individual differences related to
gender role orientation during mid to late childhood, such that at even ear-
lier stages of development the adoption of a feminine sex-role stereotype
(for females and males) puts one at risk for the expression of one's opin-
ions? These are intriguing questions that we plan to address in our future
work.
Lack of voice in the classroom

Given the focus of this volume, our findings on voice in the classroom,
around both teachers and classmates, seem particularly pertinent. As not-
ed above, in a middle school sample, voice with teacher was lower than
voice with peers and close friends, and this effect was more marked for fe-
males. In a more recent sample of adolescent girls in an all-female high
school, Johnson (1994) found that voice with teacher was lower than voice
with close friends, parents, or classmates. The only group Johnson noted
with whom voice was lower than with teachers was boys in social situa-
tions.
These findings are very timely, given observations from the AAUW re-
port (1992) and the work of the Sadkers (Sadker & Sadker, 1994) demon-
trating that girls are shortchanged in our school system. According to
these reports, girls receive less attention, less encouragement, less oppor-
tunity to develop their ideas, and less support for their abilities from
teachers in the classroom, compared to boys. From our perspective, one
logical outcome would be loss of voice around one's teachers, given the
hypothesized lack of support or validation.
However, one needs to be cautious in making generalizations about a
given gender. Our own findings reveal tremendous individual differences
within gender. As noted above, gender role orientation is a more powerful
predictor than gender per se. In Johnson's dissertation, it was hypothe-
sized that perceived support for voice would also predict self-reported lev-
el of voice. Support for voice was operationalized by items stating that
others were interested in what they had to say and encouraged them to ex-
press their opinions. In order to examine the combined effects of gender
orientation and support for voice, we first identified two groups within
our all-girls high school sample, those who could be categorized as femi-
nine and as androgynous. (The masculine group was too small to include
in these analyses.) We divided subjects within each of these two groups
into three levels of support for voice, low, medium, and high support. For
the purposes of this chapter, we will report the findings for voice with
teachers and voice with classmates (although the patterns are quite similar
across the other relationships, namely males in social situations, parents,
and close friends).
The patterns for voice with teachers and with classmates are presented
in Figure 2.5. The additive effects of gender orientation and level of sup-
port are quite clear. As in our middle school study, we find that the femi-
nine girls report considerabley lower levels of voice than do the androgy-
nous girls. Moreover, within each of these gender orientation groups, the
effects of support for voice are systematic and quite striking. Low per-
ceived support for voice results in lower levels of voice than does moder-
ate support, which in turn results in lower levels of voice than does high
support. The differences between the extreme groups are also very dra-
3.50
I I Low Teacher Support
Q Moderate Teacher Support
• High Teacher Support
2.00
FEMININE ANDROGYNOUS
4.00 T
n Low Classmate Support

•Moderate Classmate Support
• High Classmate Support
3.50
>
UJ
O 3.00
X
UJ
g
2.50 -
2.00
FEMININE ANDROGYNOUS
Figure 2.5. The impact of gender orientation and support for voice on
level of voice with teachers and classmates.
matic. Those who endorse a feminine gender role orientation, combined

with low levels of support for voice, report the lowest levels of voice (M =
2.37 with teachers and M = 2.42 with classmates). In marked contrast,
those who report an androgynous gender orientation coupled with high
levels of support for voice report extremely high levels of voice (M= 3.41
with teachers and M= 3.64 with classmates).
In this study we also obtained self-reported reasons for lack of voice.
These motives were drawn from Gilligan's theorizing, as well as from pi-
lot work in which we asked adolescents to respond to open-ended ques-
tions that asked them to indicate their reasons for not expressing their
opinion in each relational context. A factor analysis of these reasons re-
sulted in four very interpretable factors that also constitute reliable sub-
scales: (a) Lack of validation of self (others won't listen to me, won't take
me seriously, won't understand me); (b) Threats to the relationship (will
cause conflict, will lead to tension in the relationship, others will be mad
at me, they will reject or leave me); (c) Affective reactions (I'd feel embar-
rased, others would make fun of me, I'd feel stupid); and (d) Lack of opin-
ion (I don't have an opinion, I'm not sure what I think).
We examined these reasons in just those subjects with relatively low
levels of voice (those with scores of 2.60 or lower). The findings revealed
a somewhat different pattern for voice with teacher and voice with class-
mates. The primary motive for suppressing one's voice with teachers is
lack of validation. The most highly endorsed reasons for lack of voice
with classmates were lack of validation as well as negative affective reac-
tion (feeling embarrassed, stupid, tec). Subjects also endorsed lack of
opinion and threats to the relationship more highly as reasons for sup-
pressing voice with classmates than with teachers (although threats to the
relationship were endorsed even more highly with close friends and par-
ents).
Why should we be concerned about lack or loss of voice? If females
choose to adopt gender role stereotypes for which they are reinforced by
many segments of society, then what is the cause for concern? We should
be concerned since other findings (Harter et al., in press; Harter, Waters,
Pettitt, Whitesell, Kofkin, & Jordan, in press) reveal that suppressing one's
true self takes its toll on the well-being of both females and males. Our
studies reveal that among both adolescents and adults, those who do not
receive validation for the expression of their true or authentic self engage
in false self behavior, which in turn is associated with low self-esteem,
hopelessness, depressed affect, and the acknowledgment that one does not
even know one's true self. Jordan and colleagues from the Sone Center
(Jordan, 1991; Jordan, Kaplan, Miller, Stiver, & Surrey, 1991; Stiver &
Miller, 1988) also refer to the lack of zest and related depressive symp-
toms that accompany the suppression of one's authentic self. This constel-
lation in turn, limits one's ability to achieve one's potential, to make mean-
ingful contributions to society, and to be productive in one's chosen areas
of endeavor. Thus, identifying those factors that lead both females and
males to suppress their true selves is an important step in hopefully pre-
venting loss and lack of voice in children, adolescents, and adults. The
school context is a particularly powerful arena warranting intervention
with children and adolescents.
Conclusions
I began with an exploration of those factors that may serve to attenuate
students' intrinsic interest in learning, focusing on both teacher and class-
mate variables. It would appear that a constellation of factors identified by
Eccles and her colleagues are responsible for attenuating intrinsic motiva-
tion in some, but not all, students. Teachers' increased emphasis on
grades, competition, and control of students, coupled with a decreasing
personal interest in students, as well as increases in classmate social com-
parison, were specifically examined. Support was offered for a model in
which these environmental factors heighten the salience of one's academic
ability. During educational transitions, such factors also serve to cause
students to reevaluate their scholastic competence, relative to other stu-
dents. Those students who report that their competence had decreased as a
function of the transition show parallel decreases in intrinsic motivation.
Conversely, those who feel more academically competent in the new set-
ting report increases in intrinsic motivation. Those reporting no change in
their level of competence also report no change in their level of intrinsic
motivation. Further evidence reveals that students are aware of the hypoth-
esized shifts in the educational philosophy espoused by teachers as well as
of increases in social comparison. The students low in intrinsic motivation
report that the school environment is more debilitating than do those high
in intrinsic motivation.
The effects of perceived scholastic competence, as well as classmate
and teacher approval on self-esteem, as another index of adjustment were
also examined. Findings reveal that perceptions of competence are sub-
stantial predictors of self-esteem. With regard to the opinions of signifi-
cant others, both classmate support and teacher approval are predictive of
self-esteem. Findings indicate that it is the support of classmates, namely
peers in the "public" arena, which can be construed as approval from the
"generalized other," that is particularly relevant, more than the support of
close friends. Of further interest are findings revealing that those adoles-
cents who consciously endorse a looking-glass self model, asserting that
their self-esteem derives from the opinions of classmates, fare worse in
terms of greater preoccupation with approval, greater fluctuations in per-
ceived approval, greater fluctuations in self-esteem, lower levels of both
approval and self-esteem, as well as distractibility in the classroom.
Although teacher approval is somewhat less predictive than classmate
support, it is nevertheless another critical determinant. Of particular inter-
est is the finding that among students who report low parent support, the
presence of teacher support serves to increase students' level of self-es-
teem, suggesting a compensatory process. The overall pattern, with regard
to parent and teacher support, reflects an additive model in which the
highest levels of self-esteem are displayed by students reporting high par-
ent and teacher support, whereas the lowest levels of self-esteem are
found in students with both low parent and low teacher support.
Finally, factors that cause adolescents to display one form of false self
behavior, namely, suppression of their voice, were examined, as another
index of adjustment. According to Gilligan, females are particularly at
risk as they move into adolescence, when they suppress their opinions in
order to conform to societal expectations and to preserve relationships
by compromising the self. Our own findings reveal that it is those adoles-
cent females (as well as males) who endorse feminine sex-role stereotypes
that are most likely to exhibit loss of voice. Moreover, suppression of
voice is also strongly predicted by perceived support for voice. Thus, our
findings caution against generalizations that loss of voice is primarily
gender related. Our process analysis reveals that both gender orientation
and perceived support for voice are highly predictive of level of voice.
Level of voice is particularly relevant to psychological adjustment since it
is also related to self-esteem, hopelessness, and depression. Our analysis
of the antecedents of the three constructs in question, motivational orien-
tation, self-esteem, and level of voice not only caution against general-
izations but identify those variables that may be critical targets for inter-
vention.
Acknowledgments
The research reported in this chapter was supported by a grant from NIH.
The author would like to thank Steve Cohen, principal as well as the
teachers and students of Flood Middle School for their cooperation in

making these studies possible.
References
American Association for University Women's (AAUW) Report. (1992). How
schools shortchange girls. American Association of University Women Educa-
tional Foundation.
Bartholomew, K. (1990). Avoidance of intimacy: An attachment perspective.
Journal of Social and Personal Relationships, 7, 147-178.
Boldizar, J. P. (1991). Assessing sex-typing and androgeny in children: The chil-
dren's sex-role inventory. Developmental Psychology, 27, 505-515.
Brush, L. (1980). Encouraging girls in mathematics: The problem and the solu-
tion. Cambridge, MA: Abt Books.
Condry, 1, & Koslowski, B. (1979). Can education be made "intrinsically interest-
ing" to children? In L. Katz (Ed.), Current topics in early childhood education,
(vol. II). Norwood, NJ: Ablex.
Connell, J. P., & Ryan, R. M. (1984). A developmental theory of motivation in the
classroom. Teacher Education Quarterly, 11, 64-77.
Cooley, C. H. (1902). Human nature and the social order. NY: Charles Scribner's
Sons.
de Charms, R. (1968). Personal causation. New York: Academic Press.
deCharms, R. (1976). Enhancing motivation in the classroom. New York: Irving-
ton, Halsted-Wiley.
Deci, E. L. (1975). Intrinsic motivation. New York: Plenum.
Deci, E. L., & Ryan, R. M. (1985). Intrinsic motivation and self-determination in
human behavior. New York: Plenum Press.
Eccles, J., & Midgley, C. (1988). Stage-environment fit: Developmentally appro-
priate classrooms for young adolescents. In R. E. Ames & C. Ames (Eds.), Re-
search on Motivation in education, goals and cognitions (vol.3 pp. 139-186).
New York: Academic Press.
Eccles, J. S., & Midgley, C. (1990). Changes in academic motivation and self-per-
ceptions during early adolescence. In R. Montemayor, G. R. Adams, & T. P.
Gullotta (Eds.), Advances in adolescent development: From childhood to ado-
lescence (vol. 2, pp. 134-155). Newbury Park, CA: Sage.
Eccles (Parsons), J., Midgley, C, & Adler, T. F. (1984). Grade-related changes in
the school environment: Effects on achievement motivation. In J. G. Nicholls
(Ed.), The development of achievement motivation. Greenwich, CT: JAI Press.
Felson, R. B. (1993). The (somewhat) social self: How others affect self-ap-
praisals. In J. Suls (Ed.), Psychological perspectives on the self (vol. 4). Hills-
dale, NJ: Lawrence Erlbaum Associates.
Gilligan, C. (1982). In a different voice. Cambridge, MA: Harvard University
Press.
Gilligan, C, Lyons, N., & Hanmer, T. J. (1989). Making connections. Cambridge,

MA: Harvard University Press.
Gottfried, E. (1981). Grade, sex, and race differences in academic intrinsic moti-
vation. Paper presented at the Annual Meeting of the American Educational
Research Association, Los Angeles.
Harter, S. (1981). A new self-report scale of intrinsic versus extrinsic orientation
in the classroom: Motivational and informational components. Developmental
Psychology, 17, 300-312.
Harter, S. (1986). Processes underlying the construction, maintenance, and en-
hancement of the self-concept in children. In J. Suls & A. G. Greenwald (Eds.),
Psychological perspectives on the self (vol. 3, pp. 137-181). Hillsdale, NJ: Erl-
baum.
Harter, S. (1990). Causes, correlates and the functional role of global self-worth:
A life-span perspective. In J. Kolligian & R. Sternberg (Eds.), Perceptions of
competence and incompetence across the life-span (pp. 67-98). New Haven,
CT: Yale University Press.
Harter, S. (1992). The relationship between perceived competence, affect, and
motivational orientation within the classroom: Process and patterns of change.
In A. K. Boggiano & T. Pittman (Eds.), Achievement and motivation: A social-
developmental perspective. New York: Cambridge University Press.
Harter, S. (1993). Causes and consequences of low self-esteem in children and
adolescents. In R. F. Baumeister (Ed.), Self-esteem: The puzzle of low self-re-
gard. New York: Plenum.
Harter, S., & Connell, J. P. (1984). A comparison of alternative models of the rela-
tionships between academic achievement and children's perceptions of compe-
tence, control, and motivational orientation. In J. Nicholls (Ed.), The develop-
ment of achievement-related cognitions and behaviors. Greenwich, CT: JAI
Press.
Harter, S., & Jackson, B. J.(1992). Trait vs. nontrait conceptualizations of intrin-
sic/extrinsic motivational orientation, Motivation and Emotion, 16, 209-
230.
Harter, S., Whitesell, N. R., & Kowalski, P. (1992). Individual differences in the
effects of educational transitions of young adolescents' perceptions of compe-
tence and motivational orientation.
Harter, S., Marold, D. B., & Whitesell, N. R. (1991). A model of psychosocial risk
factors leading to suicidal ideation in young adolescents. Development and
Psychopathology, 4, 167-188.
Harter, S., Marold, D. B., Whitesell, N. R., & Cobbs, G. (In press). A model of the
effects of parent and peer support on adolescent false self behavior. Child De-
velopment.
Harter, S., & Monsour, A. (1992). Developmental analysis of opposing self-attri-
butes in the adolescent self-portrait. Developmental Psychology, 28, 251-260.
Harter, S., Stocker, C, & Robinson, N. (in press). The perceived directionality of
the link between approval and self-worth: The liabilities of a looking glass self
orientation among young adolescents, Journal of Adolescence.
Harter, S., Waters, P., & Gonzales, R. (1994). The ability to voice one's opinion
among middle school females and males. University of Denver, unpublished
manuscript.
Harter, S., Waters, P., Pettit, L., Whitesell, N. R., Kofkin, J., & Jordan, L. (in
press). Automony and connectedness as dimensions of adult relationship styles.
Journal of Personal and Social Relationships.
James, W. (1892). Psychology: The briefer course. NY: Henry Holt & Co.
Johnson, E. Factors influencing the self-system for students in an all-girls high-
school. University of Denver, Unpublished manuscript.
Jordan, J. (1991). The relational self: A new perspective for understanding
women's development. In G. Goethals & J. Strauss (Eds.), The self: An interdis-
ciplinary approach. New York: Springer-Verlag.
Jordan, J., Kaplan, A.G., Miller, J. B., Stiver, I. P., & Surrey, J. L. (1991). Women s
growth in connection. New York: Guilford Press.
Lepper, M. R., Greene, D., & Nisbett, R. E. (1973). Undermining children's in-
trinsic interest with extrinsic rewards: A test of the "overjustification hypothe-
sis." Journal of Personality and Social Psychology, 28, 129—137.
Mead, G. H. (1925). The genesis of the self and social control, International Jour-
nal of Ethics, 35, 251-273.
Mead, G. H. (1934). Mind, self, and society. Chicago: University of Chicago
Press.
Nicholls, J. G. (1979). Quality and equality in intellectual development: The role
of motivation in education, American Psychologist, 34, 1071-1084.
Rosenberg, M. (1986). Self-concept from middle childhood through adolesence.
In J. Suls & A. G. Greenwald (Eds.), Review, Psychological Perspective on the
self. Hillsdale, NJ: Lawrence Erlbaum Associates, pp. 182-205.
Ruble, D. N. (1983). The development of social comparison processes and their
role in achievement-related self-socialization. In E. T. Higgins, D. N. Ruble, &
W. H. Hartup (Eds.), Social cognition and social devlopment. New York: Cam-
bridge University Press.
Sadker, M., & Sadker, D. (1994). Failing at fairness: How America s schools
cheat girls. New York: Charles Scribner's Sons.
Simmons, R. G., Blyth, O. A., & Carleton-Ford, M. A. (1982). The adjustment of
early adolescents to school transitions. Paper presented at the American Re-
search Association Annual Meeting, New York.
Simmons, R. G., Rosenberg, F., & Rosenberg, M. (1973). Distrubances in the
self-image at adolescence. American Sociological Review, 38, 553-568.
Stiver, I. P., & Miller, J. B. (1988). From depression to sadness in the psychothera-
py of women. Work in progress No. 36. Wellesley, MA: Stone Center Working
Paper Series.
Self-presentation tactics promoting teacher
and peer approval: The function of excuses
and other clever explanations
Jaana Juvonen
When children first enter school, they learn a set of rules for appropriate
social conduct. Some of these rules are specific for the setting, such as:
"When you have a question you must raise your hand," or that "For lunch,
you need to line up with others." Other enforced rules capture more gener-
al guidelines for prosocial behavior. Children are told to be polite, be con-
siderate of one another, be willing to assist their classmates, and so on.
When children adopt these norms and act in a socially responsible man-
ner, they are approved by their teachers and accepted by their classmates
(see for review Wentzel, 1991). Social approval is likely to enhance their
feeling of belonging and relatedness, which in turn promotes their partici-
pation in various school activities and strengthens their identification with
the school culture (c.f. Finn, 1989). When students feel that they are part
of their school, they are also likely to be academically motivated and en-
gaged (see Hymel, Colin, Schonert-Reichl, & McDougall, this volume).
Although many important social rules and norms for acceptable behav-
ior are explicitly taught or at least are enforced by teachers (or parents) in
early elementary grades, there is another set of rules for social conduct
that also have important interpersonal as well as intrapersonal conse-
quences. These rules, or more accruately psychological principles, are
subtle and rarely (if ever) explicitly taught, yet most children seem to ac-
quire them as they interact with others. Such psychological principles per-
tain to controlling and manipulating other people's reactions toward one-
self. For example, if a child has hurt the feeling of a classmate, she needs
to know how to make amends. Although she may have been taught to
apologize in such situations, she is likely to need to do more or something
other than that to soothe the feelings of the peer. By giving an account that
portrays her act as unintentional, the child enhances the genuineness of
her expressed sorrow and enhances forgiveness (Juvonen, 1991a).
43
In this chapter, I will review recent research that my colleagues and I

have conducted on students' understanding of excuses and other self-pre-
sentation tactics that allow them to manipulate the social responses of
teachers and classmates. These social tactics reveal important aspects of
students' school adjustment in that they indicate whether students under-
stand and are sensitive to various implicit classroom norms, and can act
according to the values and expectations of those with whom they interact
in achievement situations. This research, similar to the studies described
in the other chapters of this book, illustrates why it is necessary to study
the social cognitions, interactions, and relationships of students to under-
stand their school adjustment.
The research that I review in this chapter is guided by attribution theo-
ry (Weiner, 1986). Traditionally, attributional research in the achievement
domain focused on intrapsychological phenomena. Researchers have
studied, for instance, how causal interpretations of personal achievement
failures explain different response patterns (helplessness vs. persistence)
to these negative outcomes (e.g., Dweck, 1975). It has been documented
that a perception of failure due to low aptitude (a stable and uncontrollable
cause) lowers expectation of future performance and thus creates a feeling
of hopelessness (see Weiner, 1986). In contrast, investigators have demon-
strated that failure ascribed to lack of effort, which is typically perceived
as changeable, results in maintaining expectation of success and high task
persistence (e.g., Graham, 1984). These processes capture the intraindi-
vidual approach to motivation from an attributional perspective.
Recently, attributional research in the achievement area has increasing-
ly embraced other-perceptions and the role of social emotions as motiva-
tors of behavior (Weiner, 1994). For example, Graham (1984) documents
that anger and sympathy associated with low task performance function as
attributional cues that students learn to interpret. When anger is communi-
cated, students tend to make an inference that they failed because of lack
of effort, whereas when sympathy is communicated, learners infer that
they failed because of low ability (see also Graham, 1990). The research
in this chapter extends this interpersonal attributional approach. The stud-
ies reviewed here convey how children and adolescents use their attribu-
tional knowledge to manipulate other people's social responses. Specifi-
cally, I will describe (1) how students publicly explain to others situations
in which they have broken a social contract or violated a relationship
norm, and (2) how they account for their personal achievement failures as
well as successes in ways that enhance teacher liking or facilitate peer ap-
proval.
Self-presentation tactics: The function of excuses 45
Broken social contracts: The function of excuses

When students do things they are not supposed to do or do not do things
that they should do, their teachers typically convey anger and disapproval.
For example, in a recent study I asked Finnish sixth-grade students to de-
scribe a situation when someone in their class "made the teacher very
mad" (Juvonen, 1995). Conflicts with teachers were most typically report-
ed to have involved a violation of a general expectation, rule or a norm
(e.g., not doing homework, lying to the teacher). Similarly, if students do
not follow the rules and norms that define their friendships with class-
mates, then they are likely to be confronted with negative reactions from
their peers (Hartup, 1992). Indeed, when the Finnish sixth graders were
asked to describe a situation when a classmate "made you very mad," they
conveyed that such situations most often develop when the other person
broke a promise or vow (e.g., did not show up to an engagement, did not
keep a secret) or when they hurt the other physically or psychologically
(e.g., by spreading bad rumors). Conflict situations like these involving
teachers and classmates have at least temporary negative social conse-
quences for the transgressor. Therefore, children must learn how to handle
these potentially harmful encounters.
In instances when norms or expectations are violated, the transgressor
needs to know how to reduce other's anger and decrease disapproval. For
example, a student who breaks a school rule may try to deny the behavior
to avoid punishment (cf. Schadler & Ayers-Nachamkin, 1983). Although
this strategy may work in some situations, a more developmental^ ad-
vanced tactic to reduce negative social consequences is to deny one's re-
sponsibility for the act (Snyder & Higgins, 1988; Snyder & Higgins,
Stucky, 1983; Weiner, Amrikhan, Folkes & Verette, 1987; Weiner, Fig-
uroa-Munoz, Kakihara, 1991). After all, a teacher is more likely to be an-
gry at and punish a student if he or she believes that the student did not
turn in homework because he watched TV all night (and therefore is re-
sponsible) than if the student is believed "to have gone through some
tough times because his parents are getting a divorce." This latter explana-
tion reduces responsibility because the divorce represents a mitigating cir-
cumstance (Weiner, 1995). Similarly, when a youngster does not show up
for an engagement with a classmate, the peer is more likely to be angry
and less prone to forgive the other if the transgression was because "he
had better things to do" than if "he got sick" (Juvonen, 1991a; Weiner &
Handel, 1985).
Given that perceived responsibility for negative events augments aver-
sive reactions from others, children learn to manipulate others' responses

by altering their causal explanations for events that could potentially elicit
anger, blame, and disapproval (Darby & Schlenker, 1989). Thus, in the
face of potential conflict situations, youngsters can deflect personal re-
sponsibility by using excuses that reveal the cause of the transgression
was unintentional or uncontrollable by them (e.g., "I was late because my
Mom forgot to wake me up"). In contrast, students should not readily ad-
mit that they were late on purpose or because of some controllable reason
(e.g., "I just didn't feel like getting up") given that this does not reduce
personal responsibility.
To investigate developmental differences in children's understanding
of the social consequences of "good" and "bad" excuses, Weiner and Han-
del (1985) had American 5-12 year-old children listen to short stories de-
scribing a broken social contract (e.g., "You promised to go to your
friend's house to finish up a school project but never showed up"). The
children were then given four explanations that portrayed the transgressor
as responsible (e.g., "You didn't feel like going") and four accounts that
conveyed lack of responsibility or intentionality (e.g., "Your bike got a flat
tire on the way to your friend's house") for why they did not show up.
They indicated how likely it was that they would reveal or withhold each
of these eight explanations. After each account, children also predicted
how angry their friend would be. The results documented that children in
all age groups (5-7, 8-9, and 10-12) preferred explanations that convey
lack of responsibility. Also, children perceived that such accounts elicit
less anger than do the ones that depict the person as responsible for break-
ing the social contract. These findings suggest that by 5-7 years of age,
children recognize both the principles and functions of excuse-giving.
In a similar experiment, I asked a group of second-grade Finnish stu-
dents (8-year-olds) to role play a scenario in which they had, contrary to a
prior aggreement, failed to help either their classmate (with homework be-
cause the friend had been absent from class) or their teacher (with setting
up a class exhibit after school) (Juvonen, 1995). Rather than asking them
to respond to a set of good and bad excuses, they were to produce expla-
nations that "would not make the teacher/classmate mad." The children
spontaneously provided accounts that clearly conveyed lack of intention-
ality. The most used cause was memory (e.g., "I forgot!"/"Didn't remem-
ber") when dealing with their teachers, whereas with friends they were
equally likely to implicate their forgetfulness and other commitments
(e.g., "I had to visit Grandma"). It is interesting to note that children do
not necessarily externalize the reason, as Snyder and Higgins (1988) have
proposed, but rather manipulate the perceived intentionality of the act

(such as blaming their memory).
How children's understanding and repertoire of excuses expands and
becomes more sophisticated are interesting questions to examine further.
Based on some preliminary findings, I expect that there are at least three
overlapping levels or phases that capture the development of excuse-mak-
ing:
1. Denial of the act (i.e., "I didn't do it").

2. Denial of intent (i.e., "I didn't mean to do it").
3. More sophisticated methods to reduce personal responsibility.
This can be achieved, for example, by conveying a reason that is
uncontrollable by the account giver or by presenting a mitigating
circumstance.
Besides examining the types of explanations that children are likely to

communicate, it is equally important to study students' understanding of
the affective and social consequences of excuses. In another experiment
with Finnish children, I asked sixth-grade students to predict the reactions
of the victims to classmates' explanations of their transgressions. They re-
ported that when a classmate explains to her peer that she did not show
up at the movies because of some personally controllable reason (e.g., "I
didn't feel like it"), the victim is more likely to be angry, less likely to
trust them, and the peer and the classmate are less likely to get along with
one another than when an uncontrollable reason is communicated (e.g.,
"Mom wouldn't let me go"). Similarly, they predicted that when students
tell their teacher that they did not complete the homework assignments be-
cause they were out playing with friends all night, the teacher is more an-
gry and less likely to get along with the student than if an uncontrollable
explanation was communicated (e.g., "I was sick"). Thus, the students
seemed to realize that bad excuses have not only short-term negative af-
fective consequences but that they can also have undesirable long term ef-
fects on one's relationships.
What happens then when a child is not aware of excuse-making tactics
or when a child is not concerned about the reactions of others? In a recent
study, Graham, Weiner, and Benesh-Weiner (1995) documented that one
group of children who have rather serious social adjustment problems at
school, that is, aggressive males, were less sophisticated in reasoning
about excuses and more likely to use "bad" excuses (i.e., controllable ex-
planations) than their nonaggressive peers. These findings may in part ex-
plain the reputation and the social difficulties of aggressive youngsters.

By attributing violations of norms and obligations to causes that are con-
trollable by them, they are likely to be perceived as insensitive, rude, and
defiant by their peers as well as by instructors. Furthermore, once they ac-
quire such reputations, their actions are likely to be interpreted in a nega-
tive light; that is, they are more likely to be blamed for negative encoun-
ters and less likely to be credited for prosocial actions or intentions (cf.
Hymel, 1986; Hymel, Wagner, & Butler, 1990). Hence, children who do
not understand the relationship-maintaining function of excuses or who
do not care about maintaining positive interactions with others are likely
to be confronted with more frequent conflict situations than other chil-
dren. They therefore become entrapped in a cycle of maladaptive behav-
iors that exacerbate poor interpersonal relationships (cf. Dodge, 1993; see
also Erdley, this volume).
In future research, it would be important to disentangle whether moti-
vational or social-cognitive deficits (or a combination of the two) account
for social problems of aggressive children and youngsters who for some
other reason are rejected by their peers. By doing that, appropriate inter-
vention tactics could be identified. Interventions that involve training in
excuse giving might sound somewhat questionable, yet the principles un-
derlying successful social functioning appear to be captured by this very
phenomenon. After all, effective excuse giving presumes prosocial goals
(i.e., to reduce anger, to promote positive relations) as well as rather so-
phisticated social skills to come up with appropriate strategies to accom-
plish these goals. Apparently, there is a fine line that most children learn
to recognize between "healthy" excuse-making and pathological lying. Al-
though socially skillful excuse-makers are likely to be accepted, chronic
liars tend to elicit strong negative reactions from others.
How effective excuse-giving relates to moral development is another
interesting issue to explore. For example, it may be that children at some
point are guided more by moral rules (i.e., understanding of right and
wrong) than by attributional principles. Thus, they may consider honesty
to be the primary goal and expect "telling the truth" to best facilitate posi-
tive relationships. Thus, it would be informative to compare children's
views of the social consequences of excuses and confessions.
Excuse giving thus represents only one form of impression manage-
ment. Excuses pertain to situations in which a social contract or an agree-
ment with another person is violated (cf. Scott & Lyman, 1968), and in-
volve a manipulation of the communicated reason. But what about the
self-presentational process in situations that do not involve such a social
violation? For example, what is communicated to others about the causes

of personal academic failures and successes? How can students "impress"
their teachers and peers in achievement situations? I will turn to these is-
sues next.
Personal achievement outcomes: The function

of public accounts
Starting the first day at school, students are likely to be preoccupied by
social concerns, such as: "Does my teacher like me?", "What do the other
kids think of me?". Furthermore, students try to "fit in" and act according
to the expectations and norms of a desirable peer group or of a person they
desire to befriend. To master impression management strategies that facil-
itate the approval of others, children must (a) be aware of what the other
values or expects, and (b) present themselves in a manner that is consis-
tent with the other's values or expectations (cf. Baumeister, 1982; Goff-
man, 1959; Leary & Kowalski, 1990).
Research on impression management has demonstrated that there are a
host of such strategies that people can use to gain acceptance from others
(e.g., Jones, 1964; Jones & Pittmam, 1982). Consider, for instance, con-
formity: Adolescents modify their physical appearance to match that of
popular ("cool") peers, they engage in behaviors (e.g., smoking, drug-use)
that are perceived to be highly valued by the socially powerful others, and
they adopt interests (e.g., listening to certain types of rock music) compat-
ible with those they admire (Berndt, 1979; Eckert,1989).
Rather than analyzing overt strategies to conform, I again focus on at-
tributional self-presentation tactics that are used in everyday discourse (cf.
Cody & Braaten, 1992; Turnbull, 1982). By altering causal explanations
of events, the account giver can influence others' impressions of oneself
and thus manipulate their reactions (Weary & Arkin 1981; Weiner, 1992;
Weiner et-al. 1987; Weiner et al., 1991). As Turnbull writes:
"Consider, for example, a student who does much better than anyone
else on a very difficult examination. When asked by a classmates, "How
come you did so well?" the student could answer, "It's simple. I'm
smarter than the rest of you," or "I dunno. Somehow I was just able to
figure out what the questions were likely to be, and so I was well pre-
pared." The first explanation is likely to lead to the perception that the
speaker is smug and insensitive; whereas, in contrast, by playing down
the superior performance, the second explanation both attests to the

speaker's modesty and protects others' competence. Clearly the only stu-
dent in a class who is able to infer the questions that are on the examina-
tion is a superior student. But the two ways of expressing the same infor-
mation have very different interpersonal effects." (Turnbull, 1992, 107).
Although true for all types of impression management tactics, it is es-

pecially clear that verbal accounts are audience-specific (Jones &
Pittman, 1982; Jones & Worthman, 1973; Leary & Kowalski, 1990;
Schlenker & Leary, 1982). That is, an adolescent may want to portray her-
self as lucky or "good at guessing" when confronted with classmates, but
the same explanation is not likely to be communicated to a teacher. In-
stead, a student may tell the teacher that she succeeded because she stud-
ied hard, given that teachers are known to value diligence (e.g., Weiner &
Peter, 1973).
To examine students' understanding of the social function of achieve-
ment accounts, Tamera Murdock and I conducted a series of experiments
that examined students' perceptions of hypothetical others and their self-
presentational preferences in negative and positive achievement situa-
tions. In the first experiment (Juvonen & Murdock, 1993), we examined
whether early adolescents are aware of the social implications of
achievement attributions. We asked American eighth-grade (Caucasian,
upper-middle to middle class) students to evaluate teachers' and peers'
approval of imaginary students who varied in terms of their achievement
level as well a their level of ability and effort (cf. Weiner & Kukla,
1970). We expected students to understand that the social responses of
others vary not only as function of how successful one is in school, but
also depend on why one does well or poorly. Furthermore, we presumed
that early adolescents realize that people's reactions depend also on their
values and expectations (e.g., whether they regard diligence as desirable
or not).
The results revealed that the eighth graders believed that regardless of
the achievement level, teachers prefer diligent pupils over lazy ones (cf.
Weiner & Kukla, 1970; Weiner & Peter, 1973). Whereas instructors were
perceived to most like the high achievers who are smart and diligent, these
individuals were believed to be least popular among their peers. Thus, ear-
ly adolescents appeared to be well aware of the different impact of effort
and ability ascriptions on the approval of teachers versus classmates.
In the second experiment, the same eighth grade students responded to
scenarios in which they themselves were to explain to their teachers,
classmates, and parents why they had failed or succeeded on an important

exam. We expected that their understanding of the social consequences of
effort and ability attributions would mirror their own self-presentational
preferences. For instance, we predicted that the eighth-graders would por-
tray themselves to their instructors and parents as more diligent than to
their peers. In addition to rating the likelihood of communicating effort
and ability accounts, two other typical achievement explanations, namely,
test fairness and luck, were also included. To control for the social motive
across the three audiences (i.e., their motivation to get along with each),
the participants were asked to respond in ways that would facilitate social
approval.
The results showed that the eighth-graders were more willing to con-
vey to the two adult audiences than to their classmates that they did well
because they studied hard (see Figure 3.1). By the same token, in the fail-
ure situation they were less likely to tell their teachers and parents than
their peers thay they failed because they did not study, as shown in Figure
3.1. Thus, adolescents recognized that parents as well as teachers value
Effort
Success Failure
Figure 3.1. Likelihood of communicating effort accounts as the reasons
for exam success and failure to teachers, parents, and peers. Data from
Juvonen & Murdock (1993).
diligence and thus, they wanted to portray themselves as hard workers to

elicit the adults' approval. With peers, on the other hand, they desired to
portray a different perception of themselves. They wanted to convey to
their classmates that they did not exert effort.
When dealing with parents and teachers, as opposed to classmates,
adolescents were also more likely to ascribe their positive achievement
outcomes to their competence and less inclined to attribute their success
to good luck. The high endorsement of ability and effort explanations ver-
sus low endorsement of luck accounts suggests that early adolescents be-
lieve that personal (i.e., internal) attributions increase adults' approval.
These findings are consistent with the self-presentation tactics of adults,
who tend to take credit for positive outcomes in the presence of their supe-
riors, but not when dealing with their co-workers (Pandey, 1981). Thus, it
appears that modesty rather than self-enhancement is a tactic that elicits
approval of peers, whether co-workers or classmates.
A somewhat surprising finding was obtained regarding students' pre-
ferred public accounts for failue. The early adolescents were more likely
to convey to the adults than to their classmates that their poor grades were
due to low ability. Furthermore, compared to the other accounts, the stu-
dents were rather willing to use this explanation with both their parents
and teachers. As pointed out earlier, failure due to low ability is known to
elicit sympathy in others (Graham, 1984). It appears that students in this
experiment realized they they could use this attribution-emotion link in
their favor. By attributing their failure to low ability, they not only reduced
their responsibility for the outcome (and thus decreased anger and disap-
proval of the adults) but perhaps hoped to elicit a sympathetic and nurtur-
ing response from their parents and teachers.
Although adolescents were generally more likely to convey similar rea-
sons to the two adult groups than to their classmates, there was one excep-
tion: They were less willing to tell their instructors than parents (and
peers) that their poor performance was due to the unfairness of the exam
and more likely to admit to their teachers than parents (and peers) that
they were successful because the test was fair. Attributing failure to the
test appears to negatively impact a student-teacher relationship in that
such an explanation is likely to be interpreted as an accusation. In con-
trast, by ascribing one's success to the exam, the student compliments the
teacher and thus enhances the relationship with the instructor. In summa-
ry, the results of these experiments intimate that early adolescents are both
sensitive and responsive to their audiences.
One of the most intriguing findings of the research just described was
that early adolescents desired to portray themselves to their classmates as

not putting forth effort in failure situations and downplayed the role of
hard work in the face of success. Given that these self-presentational
styles conflict with one of the fundamental values communicated at
school (i.e., the importance of effort), we wanted to further examine stu-
dents' understanding of the social meaning or value of diligence versus
laziness. We were especially interested in comprehending the development
of self-presentation tactics that might undermine one's achievement striv-
ings.
Assuming that the audience-specificity of achievement accounts re-
flects the perceived values and expectations of authority figures versus
peers, we hypothesized that there should be age-related changes in stu-
dents' self-presentational use of effort ascriptions. Given that young chil-
dren are more willing to accept the values and comply with the norms en-
dorsed by authority figures (e.g., the importance of diligence) than are
adolescents (Berndt, 1979; Steinberg & Silverberg, 1986; Youniss &
Smollar, 1985), the values and norms of their peer culture are likely to be
consistent with, or at least not oppositional to, those of adults. Hence, we
predicted that the audience specificity of students' responses should in-
crease by early adolescence. We also examined the preferences of young-
sters to publicly attribute their failures and successes to ability because
age-related changes in youngsters' preference to be diligent versus not
hard-working have been associated with their developing understanding
of the link between ability and effort (e.g., Harari & Covington, 1981;
Nicholls, 1978; Nicholls, Patashnick, & Mettetal, 1986; Chapman &
Skinner, 1989).
Adopting the methods from the earlier investigation, we instructed an
American public school sample of 4th, 6th, and 8th grade students (1/3
African-American, mixed SES) to explain to their teachers and popular
peers why they failed or succeeded in an important exam (Juvonen &
Murdock, 1995). Teachers rather than parents were used as the adult audi-
ence because instructors and classmates are both present in the context
where students' failures and successes take place. The peer group was de-
fined as those who are considered popular because popular peers are typi-
cally perceived to be the ones that others look up to, wish to be like, and
also elicit the most envy. The students were again asked to respond in
ways that would facilitate their relationships with the teacher and peers.
To examine youngsters' perceptions of the social value of the two attribu-
tions, we also included the experiment in which participants evaluate
teachers' and peers' approval of imaginary students who vary in their
achievement, ability, and effort levels. I will discuss the main findings of
the two experiments jointly by focusing first on effort accounts.
Effort explanations
We found that by fourth-grade, students can vary their explanations of
negative and positive achievement events in ways that they assume pro-
mote social approval. For example, fourth graders believed that both
teachers and popular students like those who put forth effort more than
those who are lazy. Accordingly, they wanted to portray themselves as
diligent to their teachers as well as popular classmates. Although the
eighth graders also perceived that instructors prefer effortful students to
lazy ones, they believed that popular peers are more likely to accept high
achieving classmates who do not exert effort (see dotted lines in Figure
3.2) than those who are diligent (see solid lines in Figure 3.2). (Note that
the perceived peer status of the smart and diligent students dropped from
the most popular at the fourth grade to the least popular at the eighth
grade. In contrast, the low ability, lazy students were viewed as least pop-
ular at the fourth grade but among the most popular by the eighth grade.)
Consistent with these beliefs, the eighth graders were less likely to convey
to their peers than instructors that they had done well because of effort
and more willing to tell peers than teachers that they failed because they
did not study (see Figure 3.3).
Considering the results across the failure and success conditions, a
clear grade-related difference was documented in student preferences to
communicate effort accounts to their peers versus their teachers. Further-
more, the results of our study revealed that this shift in self-portrayal in-
deed paralleled grade-level differences in the perceived effects of dili-
gence on peer popularity.
But why would early adolescents want their peers to believe that they
did not study? Given that high effort implies that the task is important to
the actor (Brown & Weiner,1984), lack of effort could be believed to com-
municate lack of importance or indifference. Inasmuch as effortful
achievement behavior is recognized as valued by teachers and parents, re-
ports of lack of effort should indicate that the student does not agree or
comply with such values. Defiance of adult norms and values, in turn,
should appeal to peers, assuming that adolescents question authority and
challenge traditional school norms (Coleman, 1961; see also Fordham &
Ogbu, 1986).
Another interpretation of the willingness of adolescents to downplay
6T
5 --
Ability, Effort
Low, High
4 -- High, High
High, Low
Low, Low
2—
4th 6th 8th

Grade
Figure 3.2. Perceived peer popularity of academically successful hypo-
thetical students across the fouth, sixth, and eighth grades. Data from
Juvonen & Murdock (1993).
the lack of effort is that this might protect the student from public humili-
ation and shame (Covington,1984). After all, unlike younger children,
adolescents believe that the relation between effort and ability is compen-
satory (Nicholls, 1978). Hence, by claiming that their failure is due to lack
of effort, they can protect their high ability perceptions (Covington,1984;
Jagacinski & Nicholls, 1992). However, if this was the case, it is unclear
why they would want to do this with their peers but not with their instruc-
tors. Also, the results regarding students' willingness to attribute their fail-
Failure: Lack of Effort Success: High Effort
9-
8-
7 Teacher
Peers
1 6- 6
Peers
5 5-
2 4H Teacher
4-
I 3H 3-
2-
4th 6th 8th 4th 6th 8th
Grade Grade
Figure 3.3. Likelihood of communicating effort accounts to teachers and peers as the reasons for exam failure and success
across the fourth, sixth, and eight grades.Data from Juvonen & Murdock (1995).
ures to low ability (discussed below under "ability explanations") do not

support this hypothesis.
Students' downplaying of effort is likely to be guided by their belief
that effort expenditure is devalued by their popular peers. It may be that
the perceptions of increased devaluing of effort in the peer group is influ-
enced, in part, by differences in the educational environments between el-
ementary and middle-schools. Certain alterations in structural features,
educational policies, and instructional practices that create a poor match
between the psychological needs of early adolescents and their education-
al environments are known to be detrimental to school adjustment (Eccles
& Midgley, 1989; Simmons & Blyth, 1987). It seems likely that some of
these changes (e.g., increased anonymity due to larger schools and com-
partmentalized classes, tracking according to performance level that pro-
motes differences rather than similarities among students, and increasing-
ly competitive grading practices) would cultivate peer norms and values,
which promote beliefs that one should not try hard or at least not admit
that one studies (Ames, 1992; Maehr & Midgley, 1991). For example,
when normative grading practices are used, a smart student who works
hard and "sets the curve" makes others look bad and is therefore likely to
be ostracized. Thus, students may not wish to work hard to excel, or at
least show that they do, because of a fear of peer rejection (cf. Brown,
1993). I will return to the educational implications of these findings after
discussing the results on perceived social meaning of ability ascriptions.
Ability explanations
In the experiment in which the ability of hypothetical students was manip-
ulated along with their effort and achievement levels, the ability was not
perceived to independently impact teacher liking or peer popularity at any
grade level. However, the eighth-graders were most likely to tell their
teacher that their poor grades were due to low ability in the subject matter.
These results replicate our earlier findings, yet they are inconsistent with
much of self-presentation research suggesting that people do not want to
portray themselves to others in ways that question their ability or compe-
tence, especially when seeking social approval (Covington & Omelich,
1979; Jagacinski & Nicholls, 1990; Snyder & Higgins, 1988). However,
as I indicated earlier, the present results are sensible from an attributional
perspective. Compared to young elementary school age children, adoles-
cents are known to view ability as an uncontrollable cause (Nicholls,
1978). Uncontrollable causes elicit sympathy from others (Graham, Dou-
bleday, & Guarino, 1984) which, in turn, facilitates prosocial behaviors

and social acceptance (Graham & Barker, 1990; Juvonen, 1991b; Weiner,
1986). Thus, the present data intimate that in unsuccessful achievement
situations, there may be social benefits in portraying oneself as incapable
or low in ability in a particular subject or task. Students might believe that
they can avoid punitive social responses and instead elicit approval, espe-
cially from their teachers. Using low ability as a self-presentation tactic
requires rather sophisticated understanding of practices that favorably im-
pact the reactions of others. This may also explain why only the oldest
participants were likely to use low ability to account for their achievement
failures.
In summary, the cross-sectional study showed that audience-specificity
of presenting oneself to others as diligent or low in ability increased with
advancing grade-level. These findings suggest that students recognize the
different values and expectations of their audiences and can modify their
public explanations of positive and negative events accordingly. Thus, in
contrast to the rather simple principle of excuse giving (i.e., reducing re-
sponsibility to avoid anger and disapproval), the strategies for seeking so-
cial approval are more sophisticated. Based on these experiments, a few
specific strategies that enhance the approval of significant adults and peers
can be identified. For example, it appears that in positive achievement situ-
ations, middle school students (i.e., early adolescents) are able to facilitate
social approval by (1) eliciting praise from those in the position of authori-
ty by portraying themselves as personally accountable (i.e., by using inter-
nal attributions) for their achievements, (2) complimenting their instructor
by crediting her or him for their successful achievements, and (3) appearing
modest to peers by downplaying the role of personal (cf. internal) achieve-
ment accounts. On the other hand, in the face of failure, early adolescents
can earn social approval by: (1) Denying personal responsibility for
achievement failures when dealing with superiors (e.g., teachers or par-
ents); (2) eliciting sympathy from superiors by providing accounts that are
personally uncontrollable; and (3) accepting responsibility for personal
failures when dealing with others of similar or equal status (i.e., peers).
In a more recent set of experiments with Finnish students, I expanded
the analyses of audience specificity of achievement explanations to in-
clude positive and negative relationships (Juvonen, 1994). That is, stu-
dents were asked to respond to liked versus disliked teachers and liked
versus disliked classmates. The results showed that preadolescents can use
their knowledge of others' values and expectations in ways that communi-
cate their negative sentiment toward another. For example, students were
more likely to convey to the disliked teacher than the liked instructor that
they failed because the exam was unfair. Thus, it seems that early adoles-
cents realize the relational implications of public accounts. They can use
achievement explanations not only to please people but also to aggravate
those they dislike.
I am currently examining whether these attributional self-presentation
tactics are related to the social status of students. One would expect that
youth who do not vary their achievement accounts according to the values
and expectations of their audiences are not socially adjusted. That is, stu-
dents who continue to portray themselves as diligent to their instructors as
well as to their classmates in middle-school might be considered as
"nerdy" or labelled as "teachers' pets" and thus are unlikely to be accepted
by the majority of their classmates. Feelings of lack of acceptance and
support from peers, in turn, are likely to color these students' school expe-
riences (see Birch and Ladd, in this volume; Kupersmidt, Buchele, Voe-
gler, & Sedikides, in this volume). Those who are ridiculed or "put down"
by others tend to feel lonely and experience lack of support (Cassidy &
Asher, 1992; Juvonen, Murdock, Curran, & Gordon, 1994).
In additon to affecting social approval, one would expect that the pro-
ductivity of the individual who is not accepted by her or his peers (cf.
coworkers) suffers in such conditions. To investigate how self-presenta-
tion tactics in achievement situations may be related to actual achievement
behavior is another area of research that we are beginning to explore. Al-
though I do not have any data to present at this point, let me discuss possi-
ble implications of some of our findings on achievement.
Self-presentation tactics and achievement

Given that effort expenditure and diligence are considered important in-
dices of achievement motivation, it is important to investigate ways in
which public portrayals of lack of effort and laziness might affect one's
actual achievement. First of all, youngsters (early adolescents) may be
quite well aware of the function and meaning of self-presentational tac-
tics, such as downplaying the role of effort when interacting with their
peers. They may promote this public image of themselves, yet privately
value the importance of effort and study diligently at home. In Harter's
(1990; this volume) terms, these youth would be likely to ascribe this
somewhat "schizophrenic" behavior to their "false self." In these cases,
one would not expect to find any relation between students' social and
academic status.
However, impression management research on adults indicates that

public behaviors are more committing than private behaviors or thoughts
because they force individuals to build a reputation by which they are
known and treated (McKillop, Berzonsky, & Schlenker, 1992). As dis-
cussed in the context of excuses of aggressive youngsters, public images
are difficult to revoke (Goffman, 1959; Hymel, et al., 1990; Tedeschi &
Norman, 1985; McKillop, et al. 1992). For example, a "cool kid" who has
a reputation of being uninvolved in school work cannot easily change his
or her achievement behavior without it affecting his or her social status in
the classroom. Thus, teachers and peers alike may treat such a student in
ways that become self-fulfilling prophesies. Hence, in some cases, social
success among peers may be gained at the expense of academic failure.
Also, some youngsters may not necessarily distinguish their own iden-
tity from their reflected appraisals (Harter, 1990, this volume). They may
be primarily concerned about earning approval of their peers and, hence,
adopt (and eventually internalize) values and behavior patterns that are
consistent with those of their friends (Berndt & Keefe, 1992) or peers to
whom they look up (Brown, 1993). In other words, students may not only
claim that they do not study, but also reduce their actual achievement ef-
forts in their search for peer approval. Indeed, it is during early adoles-
cence when students' grades, interest in school work, and self-concept de-
cline (see Eccles & Midgley, 1989; Simmons & Blyth, 1987 for reviews).
It would therefore be important to investigate whether peer group norms
can be altered in ways that promote effort expenditure. For example, de-
creased competition and endorsement of small group cooperation might
change the social meaning of effort (cf. Ames, 1992; Elliott & Dweck,
1988). Instead of fostering negative interdependence, effort could be con-
strued as a prosocial force that would unite peers with a common achieve-
ment goal.
Similarly to lack of effort, low ability accounts might have certain im-
plications for students' actual academic performance. Although we found
that only the oldest students were likely to communicate low ability ac-
counts, I assume this may not necessarily capture an age-related phenom-
enon but rather a behavior that is promoted by certain experiences. Based
on anecdotal evidence, I expect that compared to typical children, stu-
dents who have confronted frequent failure situations are more likely to
use low ability accounts to elicit sympathy and to lower the expectations
of their instructors. In working with children with disabilities, I have no-
ticed that they are quite clever in manipulating the social responses of oth-
ers (especially of adults who do not know them well). They might tell a
substitute teacher that their eyes hurt when asked to read a passage or that
they cannot produce better handwriting because they are not able to prop-
erly hold their pencil. In other words, some children might use their dis-
abilities to persuade the instructor to reduce demands or lower expecta-
tions. This type of self-handicapping behavior appears to be learned, and
hence experience rather than age (or developmental stage) would explain
the prevalence of it. Obviously, such self-presentational tactics might also
affect students' actual performance, either through other communicated
low expectations or via internalization of the public self.
Conclusion
As stated in the beginning of this chapter, my goal was to demonstrate that
students' choices of excuses and publicly communicated achievement ex-
planations show how their understanding of naive psychology (e.g., attri-
bution-emotion-behavior links) enables them to effectively use public ac-
counts to suit their social goals. This research also depicts the complexity
of students' social life in school. Obviously, many more questions remain
to be investigated. For example, which other type of self-presentation tac-
tics are used by students? What happens when middle school students
want to simultaneously please their instructor as well as a "cool" group of
classmates? Or what do students opt to do when their own achievement
values conflict with those of their friends? The multiple and complex so-
cial contexts that influence children's and adolescents' behavior in school
provide rich data that researchers have just begun to explore.
Acknowledgment
This research was supported by the National Academy of Education
Spencer Post-Doctoral Fellowship.
References
Ames, C. (1992). Classrooms: Goals, structures, and student motivation. Journal
of Educational Psychology, 84, 261—271.
Baumeister, R. F. (1982). A self-presentational view of social phenomena. Psy-
chological Bulletin, 91, 3-26.
Berndt, T. J. (1979). Developmental changes in conformity to peers and parents.
Developmental Psychology, 15, 608-616.
Berndt, T. J., & Keefe, K. (1992). Friends' influence on adolescents' perceptions

of themselves at school. In D. H. Schunk and J. L. Meece (Eds.), Student per-
ceptions in the clssroom. Hillsdale, NJ: Erlbaum.
Brown, B. B. (1993). School culture, social policies, and the academic motivation
of U.S. students. In T. M. Tomlinson (Ed.), Motivating students to learn: Over-
coming barriers to high achievement (pp. 63-98). Berkeley: McCutchan.
Cassidy, J., & Asher, S. R. (1992). Loneliness and peer relations in young chil-
dren. Child Development, 63, 350-365.
Chapman, M., & Skinner, E. A. (1989). Children's agency beliefs, cognitive per-
formance, and conceptions of effort and ability: Individual and developmental
differences. Child Development, 60, 1229-1238.
Cody, M. J., & Braaten, D. O. (1992). The social-interactive aspects of account-
giving. In M. L. McLaughlin, M. J. Cody, & S. J. Read (Eds.), Explaining one's
self to others: Reason-giving in a social context (pp. 225-244). Hillsdale, NJ:
Erlbaum.
Coleman, J. S. (1961). The adolescent society. New York: Free Press.
Covington, M. V (1984). The motive for self-worth. In C. Ames & R. Ames (Eds.)
Research on motivation in education (vol. 1, pp. 77-113). San Diego: Academ-
ic Press.
Covington, M. V, & Omelich, C. L. (1979). Effort: The double-edged sword in
school achievement. Jounal of Educational Psychology, 77, 446—459.
Darby, B., & Schlenker, B. (1989). Children's reactions to transgressions: Effects
of the actor's apology, reputation, and remorse. British Journal of Social Psy-
chology, 28, 353-364.
Dodge, K. (1993). Social-cognitive mechanisms in the development of conduct
disorder and depression. In L. Porter & M. Rosenzweig (Eds.), Annual review
ofpsycholgy (vol. 44 pp.559-584). Palo Alto, CA: Annual Reviews.
Dweck, C. S. (1975). The role of expectations and attributions in the alleviation of
learned helplessness. Journal of Personality and Social Psychology Journal,
25, 109-116.
Eccles, J. S., & Midgley, C. (1989). Stage-environment fit: Developmentally ap-
propriate classrooms for early adolescents. In R. Ames & C. Ames (Eds.), Re-
search on motivation in education (vol. 3, pp. 139-186). New York: Academic
Press.
Eckert, P. (1989). Jocks and burnouts: Social categories and identity in the high
school New York: Teachers College, Columbia University.
Elliott, E. S., & Dweck, C. S. (1988). Goals: An approach to motivation and
achievement. Journal of Personality and Social Psychology, 25, 109-116.
Finn, J. (1989). Withdrawing from school. Review of Educational Research, 59,
117-142.
Fordham, S., & Ogbu, J. U. (1986). Black students' school success: Coping with
"burden of'acting white.'" The Urban Review, 18, 176-206.
Goffman, E. (1959). The presentation of self in everyday life. Garden City, NY:
Doubleday Anchor.
Graham, S. (1984). Communicated sympathy and anger to black and white chil-
dren: The cognitive (attributional) consequences of affective cues. Journal of
Personality and Social Psychology, 47, 40-54.
Graham, S. (1990). Communicating low ability in the classroom: Bad things good
teachers sometimes do. In S. Graham and V S. Folkes (Eds.), Attribution theo-
ry: Applications to achievement, mental health, and interpersonal conflict (pp.
17-36). Hillsdale, NJ: Erlbaum.
Graham, S., & Barker, G. P. (1990). The down side of help: An attributional-de-
velopmental analysis of helping behavior as a low ability cue. Journal of Edu-
cational Psychology, 82, 7-14.
Graham, S., Doubleday, C, & Guarino, P. A. (1984). The development of relations
between perceived controllability and the emotions of pity, anger, and guilt.
Child Development, 55, 561-565.
Graham, S., Weiner, B., & Benesh-Weiner, M. (1995). An attributional analysis of
the development of excuse giving in aggressive and nonaggressive African-
American boys. Developmental Psychology, 31, 274-284.
Harari, O., & Covington, M. V (1981). Reactions to achievement behavior from a
teacher and student perspective: A developmental analysis. American Educa-
tional Research Journal, 18, 15-28.
Harter, S. (1990). Self and identity development. In S. S. Feldman & G. R. Elliott
(Eds.), At the threshold: The developing adolescent (pp. 352-387). Cambridge,
MA: Harvard University.
Hartup, W. W. (1992). Conflict and friendship relations. In C. U. Shanz & W. W.
Hartup (Eds.), Conflict in child and adolescent development (pp. 186-215).
Cambridge, England: Cambridge University Press.
Hymel, S., Wagner, E., & Butler, L. J. (1990). Reputational bias: Views from the
peer group. In S. R. Asher and J. D. Coie (Eds.), Peer rejection in childhood
(pp. 156-186). Cambridge, England: Cambridge University Press.
Jagacinski, C. M., & Nicholls, J. G. (1990). Reducing effort to project perceived
ability: "They'd do it but I wouldn't." Journal of Educational Psychology, 82,
15-21.
Jones, E. E., & Pittman, T. S. (1982). Toward a general theory of strategic self-
presentation. In J. Suls (Ed.), Psychological perspectives on self (vol. 1, pp.
Jones, E. E., & Worthman, C. (1973). Ingratiation: An attributional approach.
Morristown, NJ: General Learning Press.
Juvonen, J. (1991a). The effect of attributions and interpersonal attitudes on so-
cial emotions and behavior. Paper presented at the annual meeting of the Amer-
ican Educational Research Association, Chicago.
Juvonen, J. (1991b). Deviance, perceived responsibility, and negative peer reac-
tions. Developmental Psychology, 27, 672-681.
Juvonen, J. (1994). 'Why did you fail? " - Depends who s asking!" Paper present-
ed at the biennial meeting of the Society for Research on Adolescence, San
Diego.
Juvonen, J., (1995). A developmental analysis of excuses as relationship main-

taining strategies. Unpublished manuscript.
Juvonen, J., & Murdock, T. B. (1993). How to promote social approval: The ef-
fects of audience and outcome on publicly communicated attributions. Journal
of Educational Psychology, 85, 365-376.
Juvonen, J., & Murdock, T. B. (1995). Grade-level differences in the social value
of effort: Implications for self-presentation tactics of early adolescents. Child
Development, 66, 1694-1705.
Juvonen, J., Murdock, T. B., Curran, A., & Gordon, T. (1994). Age-related
changes in peer attitudes, perceived social support, and loneliness of bullies
and victims. Paper presented at the Annual Meeting of the American Educa-
tional Research Association, New Orleans, 1994.
Leary, M. R., & Kowalski, R. M. (1990). Impression management: A literature re-
view and two-component model. Psychological Bulletin, 107, 34-47.
Maehr, M. L., & Midgley, C. (1991). Enhancing student motivation: A school-
wide approach. Educational Psychologist, 26, 399-426.
McKillop, K. I, Berzonsky, M. D., & Schlenker, B. R. (1992). The impact of self-
presentations on self-beliefs: Effects of social identity and self-presentation.
Journal of Personality, 60, 789-808.
Nicholls, J. G. (1978). The development of concepts of effort and ability, percep-
tions of own attainment, and the understanding that difficult tasks require more
ability. Child Development, 49, 800-814.
Nicholls, J. G., Patashnick, M., & Mettetal, G. (1986). Conceptions of ability and
intelligence. Child Development, 57, 636-645.
Pandey, J. (1981). A note on social power through ingratiation among workers.
Journal of Occupational Psychology, 54, 65-67.
Schadler, M., & Ayers-Nachamkin, B. (1983). The development of excuse-mak-
ing. In C. R. Snyder, R. L. Higgins & R. J. Stucky, Excuses: Masquerades in
search ofgrace. New York: Wiley.
Schlenker, B. R., & Leary, M. R. (1982). Audiences' reactions to self-enhancing,
self-denigrating, and accurate self-presentations. Journal of Experimental So-
cial Psychology 18, 89-104.
Scott, M., & Lyman, S. (1968). Accounts. American Sociological Review, 23,
46,62.
Simmons, R. G., & Blyth, D. A. (1987). Moving into adolescence. Hawthorne,
NY: de Gruyter.
Snyder, C. R., & Higgins, R. L. (1988). Excuses: Their effective role in negotia-
tion of reality. Psychological Bulletin, 104, 23-35.
Snyder, C. R., & Higgins, R. L., & Stucky, R. J. (1983). Excuses: Masquerades in
search ofgrace. New York: Wiley.
Steinberg, L., & Silverberg, S. B. (1986). The vicissitudes of autonomy in early
adolescence. Child Development, 57, 841-851.
Tedeschi, J. T., & Norman, N. (1985). Social power, self-presentation, and the self.
In B. R. Schlenker (Ed.), The self and social life (pp. 293-322). New York: Mc-
Graw Hill.
Turnbull, W. (1992). A conversation approach to explanation, with emphasis on
politeness and accounting. In M. L. McLaughlin, M. J. Cody, & S. J. Read
(Eds.), Explaining one's self to others: Reason-giving in a social context (pp.
Weary, G., & Arkin, R. M. (1981). Attributional self-presentation. In J. H. Harvey,
W. J. Ickes, & R. I. Kidd (Eds.), New directions in attribution research. Hills-
dale, NJ: Erlbaum.
Weiner, B. (1986). An attributional theory of motivation. New York: Springer-Ver-
lag.
Weiner, B. (1992). Excuses in everyday interaction. In M. L. McLaughlin, M. J.
Cody, & S. J. Read (Eds. ), Explaining one's self to others: Reason-giving in a
social context (pp. 131-146). Hillsdale, NJ: Erlbaum.
Weiner, B. (1994). Ability versus effort revisited: The moral determinants of
achievement evaluation and achievement as a moral system. Educational Psy-
chologist, 29, 163-172.
Weiner, B. (1995). Judgments ofresponsibilty: A foundation for a theory of social
conduct. New York: Guilford.
Weiner, B., Amrikhan, J., Folkes, V S., & Verette, J. (1987). An attributional
analysis of excuse giving: Studies of naive theory of emotion. Journal of Per-
sonality and Social Psychology, 52, 316-324.
Weiner, B., Figuroa-Munoz, A. & Kakihara, C. (1991). The goals of excuses and
communication strategies related to causal perceptions. Personality and Social
Psychology Bulletin, 17, 4-13.
Weiner, B. & Handel, S. (1985). Anticipated emotional consequences of causal
communications and reported communication strategy. Developmental Psy-
chology, 21, 102-107.
Weiner, B. & Kukla, A. (1970). An attributional analysis of achievement motiva-
tions. Journal of Personality and Social Psychology, 15, 1-20.
Weiner, B., & Peter, N. (1973). A cognitive-developmental analysis of achieve-
ment and moral judgments. Developmental Psychology, 9, 290-309.
Wentzel, K. R. (1991). Social Competence at school: Relation between social re-
sponsibility and academic achievement. Review of Educational Research, 61,
1-24.
Youniss, J., & Smollar, J. (1985). Adolescent relations with mothers, fathers, and
friends. Chicago: University of Chicago.
Social self-discrepancy: A theory relating peer
relations problems and school maladjustment
Janis B. Kupersmidt, Kathy S. Buchele, Mary Ellen Voegler,

and Constantine Sedikides
A growing body of literature indicates that problematic peer relations in

childhood and adolescence are predictive of both academic and behavioral
problems in school (see Kohlberg, LaCrosse, & Ricks, 1972; Kupersmidt,
Coie, & Dodge, 1990; Parker & Asher, 1987, for reviews). The findings
from this literature have been important for the development of criteria for
defining, identifying, and screening children at risk for school maladjust-
ment. Despite the fact that the relation between problematic peer relations
and school maladjustment has been replicated across samples of children
from different ethnic, geographic, and developmental groups, little is
known about the mechanisms by which problematic peer relations may af-
fect school maladjustment. The development and testing of such mecha-
nisms has been slow, in part, because the transactional relations among so-
cial, behavioral, and academic functioning over time have been difficult to
study due to methodological limitations and practical constraints. It is
likely that there is a bidirectional influence between peer relations and
school maladjustment; however, this chapter focuses only on one side of
the equation, namely, the influence of problematic peer relations on
school maladjustment. In this chapter, school maladjustment is defined in
terms of behavioral, emotional, or academic difficulties that may interfere
with adequate functioning in school.
One mechanism that has been proposed for understanding the impact
that problematic peer relations have on children's school maladjustment
suggests that poor peer relations are stressful for children due to both the
experienced negative affect and the accompanying lack of social support
that, in turn, increases children's vulnerability to other life stressors (Ar-
macost, 1989; Colton & Gore, 1991; see also Birch & Ladd, this volume).
In fact, children report that being disliked or picked on by peers is a major
stressor (Jones & Fiore, 1991) and they rate negative changes in peer ac-
ceptance as equally stressful as other life events, such as failure of a year
66
Social self-discrepancy 67
in school or hospitalization of a parent for a serious illness (Johnson,

1988). Children with poor peer relations may be socially isolated or ostra-
cized and may, in turn, isolate themselves from peers by not attending
school to avoid further exposure to stress. If the social isolation of chil-
dren is associated with teasing or peer victimization, they may also devel-
op heightened social anxiety and fear of public performances in school,
such as reading aloud.
Although problematic peer relations can have negative consequences
in terms of mental health and school maladjustment, these consequences
do not necessarily generalize to all children. There may be children for
whom problematic peer relations do not affect adjustment. For these chil-
dren, the level of need for relatedness, affiliation, or peer status may be
relatively low. For example, they may describe themselves as not caring
what others think about them or not wanting or needing friends. Alterna-
tively, these children may have other means of satisfying their social
needs, such as a positive affiliation with a pet, parent, or sibling. Thus, we
need to know a child's desired type, quality, and quantity of peer relations
in order to be able to predict his or her reaction to different kinds of inter-
personal stressors with peers.
In this chapter, we propose a model for understanding the differential
impact of peer relations problems on children's school maladjustment.
Specifically, we propose that individual differences in cognitions about
the social self may account for this differential impact (see also Erdley, Ju-
vonen, this volume). The aim of the present chapter is to begin to develop
a framework for explaining individual differences in the impact of the
type, quality, and quantity of chidren's peer relations problems on school
maladjustment. We will develop a theory about the ways in which social
self cognitions mediate the relation between problematic peer relations
and school maladjustment. In addition, we will present the preliminary re-
sults of focus group discussions that are relevant to our theory.
In order to begin to examine the proposed model, we conducted focus
group discussions with 3rd, 5th, 7th, 9th, and 1 lth grade students. Trained
adult interviewers conducted separate groups with boys and girls. The old-
er groups (7th, 9th, and 1 lth grades) consisted only of African-American,
inner-city children, whereas the younger groups (3rd and 5th grades) con-
sisted of both Caucasian and African-American children. We conducted
all groups in school or a school-based summer camp during the day. The
interviewers were trained to use a semi-structured interview and were
guided by a standard script of questions used with each group of students.
Children were asked to discuss five types of peer relations including so-
rial networks, best friends, boyfriends and girlfriends, popularity, and be-
ing disliked or rejected by peers. The results of these focus group discus-
sions will be presented throughout the chapter to complement the existing
empirical literature.
The specific organization of the chapter is as follows. In the next sec-
tion, we will briefly discuss the importance of positive peer relations for
normal development, and we will provide a brief review of the literature
on the prediction of school maladjustment from different kinds of peer re-
lations problems. In the following section, we will discuss the existing
adult literature relevant to our theory. Because mediators of problematic
peer relations on school maladjustment have not been directly examined
in the child literature, we reviewed the adult literature for ideas about so-
cial cognitive mediators of interpersonal stess on maladjustment. This
review provides the foundation for the remainder of the chapter, which be-
gins with a discussion of the concept of social needs followed by sections
on a working definition of social needs and the domains of social needs.
We then review group and individual differences in social needs. Finally,
we will introduce self-discrepancy theory as a potential framework for un-
derstanding the impact of interpersonal stress, more broadly, and peer re-
lations problems, more specifically, on maladjustment. In the final sec-
tion, we will discuss various mechanisms through which cognitions about
the social self in the context of problematic peer relations may affect
school maladjustment.
The importance of peer relations for school maladjustment

The importance of establishing and maintaining healthy peer relations in
normal social development has been well-documented. Vandell and
Mueller (1980) note that in the first six months of life, children begin to
show evidence of interest in peers, such as touching, peer-directed smil-
ing, and babbling. Social interactions increase in frequency during the
preschool years (Rubin, Fein, & Vandenberg, 1983). The presence of reci-
procated friendships provides the opportunity for modeling and corrective
feedback in the social skill development of school-aged children (Hartup,
1989). Relationships with peers become increasingly important to chil-
dren, and children report that the functions these relationships serve grow
to equal or sometimes surpass relationships with other central network
members (Furman & Buhrmester, 1985).
The following comments from focus group discussions suggest the im-
portance of peer relations to children and adolescents: "I come to school
to get the gossip." "[I like] the social life. Just to see people. That's why I
like school." When asked what they like most about school, children's
most common response was "coming to see my friends." Thus, not only
are peer relations subjectively important to children, but they also appear
to motivate them to attend school.
Most of the developmental psychopathology literature in the last
15-20 years that has examined the role of peer relations has focused upon
children's relations with same-aged peers as the primary indicator of prob-
lematic peer relations. The following section discusses five types of peer
problems that may be associated with increased risk for school maladjust-
ment and how each type of problem relates to other types of peer prob-
lems. We will also discuss, when possible, individual variation in response
to exposure to each particualr peer problem.
One way that peer problems have been operationalized is through the
use of peer nomination methods to identify "rejected children," or chil-
dren who are actively disliked by a substantial number of peers while si-
multaneously being most liked by few, if any, peers (Coie, Dodge, & Cop-
potelli, 1982). This dislike or negative affect directed towards certain
peers has been found to be correlated with negative behavior being direct-
ed toward rejected children as well. Rejected children are often the targets
of overt aggression (Coie & Kupersmidt, 1983; Dodge, Coie, & Brakke,
1982) as well as the victims of relational aggression (Crick & Grotpeter,
1994).
Rejected children have been found to be at heightened risk for a wide
range of school-related problems including absenteeism, school dropout,
low academic achievement, poor grades, and grade retention (Coie,
Lochman, Terry, & Hyman, 1992; DeRosier, Kupersmidt, & Patterson,
1994; Kupersmidt & Coie, 1990; Ollendick, Weist, Borden, & Green,
1992; Wentzel, 1991; see also Hymel et al., this volume). In tests of the ef-
fects of the chronicity and proximity of peer rejection on behavioral and
academic adjustment, even one experience of peer rejection was predic-
tive of increases in school absenteeism, after controlling for prior levels of
absenteeism in the model (DeRosier et al., 1994). Likewise, the interac-
tion between the chronicity of peer rejection and prior levels of aggression
was associated with heightened levels of aggression in school. The
strength of the correlation between prior aggression and later aggression
was directly related to the chronicity of peer rejection. The academic
achievement of children in early elementary school was more negatively
affected by chronic peer rejection than the achievement of children who
experienced chronic peer rejection later in elementary school (DeRosier et
al., 1994). In addition to the impact of peer rejection on behavioral and

academic functioning, rejection has also been associate with internalizing
problems including anxiety and depression (Asarnow, 1988; French &
Waas, 1985; Waas, 1987).
In addition to these more recent findings on peer rejection and its con-
sequences, an earlier literature on unpopularity also suggests an associa-
tion with school-related problems (see Parker & Asher, 1987, for a re-
view); however, the operational definition of unpopularity varied
substantially across these earlier studies. Despite the consistent findings
regarding negative outcomes associated with peer rejection, the specific
form of negative outcome appears to vary across individuals, leading
some to suggest that peer rejection may operate as a general stressor
rather than a specific precursor to a specific disorder (Kupersmidt &
Coie, 1990; Kupersmidt & Patterson, 1991). In addition, substantial indi-
vidual differences in the response to peer rejection have been observed.
For example, Kupersmidt and Coie (1990) reported that approximately
35% of rejected children dropped out of school.
More recently, interest in the normal development of dyadic peer rela-
tionships among children has emerged in the theoretical and empirical lit-
erature (Bukowski & Hoza, 1989; Parker & Asher, 1993). Based upon this
line of research, a second way that having a peer relations problem has
been operationalized is the absence of a best friendship using reciprocal
peer nominations. Although group acceptance and friendship are consid-
ered to be distinct constructs, some empirical evidence has emerged that
suggests that these constructs are related. For example, low status children
(e.g., rejected or neglected) are less likely than other children to have a
group of prosocial friends or a best friend and are more likely than other
children to have low or average status friends (Kupersmidt, DeRosier, &
Patterson, 1995; Kupersmidt, Griesler, & Patterson, 1994; Parker & Ash-
er, 1993). Likewise, higher status popular or controversial children are
more likely than other children to have friends and have higher status
friends. However, it is noteworthy that some low status children do have
both reciprocated friendships and best friends (Kupersmidt et al., 1995;
Kupersmidt, Griesler, & Patterson, 1994; Parker & Asher, 1993). Never-
theless, low status children report less caring, instrumental aid, intimacy,
and more conflict and betrayal in their reciprocated best friendships than
higher accepted peers (Parker & Asher, 1993).
Although it has been shown that dyadic relationships, particularly best
friendships, are important to children and adolescents (Furman &
Buhrmester, 1985), little is known about the consequences for school be-
havior and performance of not having a friend. One exception is from a

recent cross-sectional study by Parker and Asher (1993), in which they re-
ported that not having a reciprocated best friend and low group accep-
tance made separate contributions to the prediction of loneliness. Thus,
this finding presents new evidence to suggest an additive relation between
two peer problems as independent risk factors.
A third peer relations problem has been defined through ratings or ob-
servations of the quality of the best friend relationship. Although many
low-accepted children have friendships, they report less caring, instru-
mental aid, and intimacy, and more conflict and betrayal in their recipro-
cated best friendships than do higher-accepted children. Some have sug-
gested that having a poor quality relationship with a best friend such as
one characterized by frequent conflicts may have negative effets on chil-
dren's development (Berndt, 1989; Rook, 1984). In particular, Berndt
(1992) reported that adolescents with more supportive friendships were
better adjusted to school than adolescents with less supportive friendships.
In addition, Goodyer, Wright, and Altham (1989) reported that friendship
difficulties were associated with both anxiety and depression. Finally,
Parker and Asher (1993) reported that various friendship qualities made
independent contributions to the prediction of loneliness, even after con-
trolling for peer acceptance.
A fourth kind of peer relations problem concerns the lack of member-
ship in a social network. The relations among low peer status, lack of a
best friend, and lack of a social network have not been thoroughly investi-
gated, so that the prevalence of the co-occurrence of these problems is not
known. In addition, the short- and long-term consequences of this type of
social isolation have not been well investigated.
A fifth kind of peer relations problem concerns the characteristics of
the child's best friend or peer network members. Members of children's
peer groups have been found to be similar with respect to their level of
motivation in school (Kindermann, 1993; this volume) and although there
is substantial turnover in the membership of groups, groups remain fairly
stable in their motivational composition. In other words, according to Kin-
dermann (1993) children who were very motivated in school tended to af-
filiate with one another, and those who were less motivated also tended to
hang out together. Similarly, Tesser, Campbell, and Smith (1984) reported
homogeneity among affiliated peers with respect to school performance
and Cairns, Cairns, Neckerman, Gest, and Gariepy (1988) reported that
aggressive children tend to associate with one another in antisocial net-
works. In addition, adolescents who are friends and, in particular, those
who stay friends across time tend to become more similar to one another
in behavior, attitudes, and school motivation (Berndt, Laychak, & Park,
1990; Epstein, 1983b; Kandel, 1978a, 1978b). Adolescents also report
substantial peer group influence that affects many aspects of life including
their appearance and illicit behaviors and cognitions (O'Brien & Bierman,
1988). Social affilation with an antisocial peer group is a powerful predic-
tor of a range of school-related problems and antisocial behaviors (Dish-
ion & Loeber, 1985; Elliot, Huizinga, & Ageton, 1985; Patterson & Dish-
ion, 1985).
A recent longitudinal study examined developmental patterns of five
indices of problematic peer relations including low group acceptance, not
having a reciprocated best friend, low perceived social support from the
best friend, high conflict with the best friend, and having aggressive
friends as predictors of aggression and delinquency (Kupersmidt, Burchi-
nal, & Patterson, 1995). Multiple peer relations problems were additively
associated with the prediction of each outcome, suggesting that children
with more problems were at higher risk than children with fewer prob-
lems. In this way, these findings are consistent with those reported earlier
for the prediction of loneliness by Parker and Asher (1993), suggesting
that the evaluation of social risk factors requires the assessment of multi-
ple aspects of social functioning with peers. In addition, these findings
provide additonal evidence for substantial individual differences in the
impact of each peer relations problem on a variety of negative outcomes.
In summary, five different kinds of peer relations problems have been
identified that are expected to be associated with heightened risk for
school maladjustment, including being rejected, not having a best friend,
having a poor quality friendship, lack of membership in a social network,
and being a member of an antisocial network. Individual differences in the
negative impact of these problems on school maladjustment were reported
in thes studies. Additional research on each of these peer relations prob-
lems is needed in several areas. First, the prevalence and interrelations
among different kinds of peer relations problems need to be studied across
development. Second, the prediction of school maladjustment from both
independent peer relations problems as well as from multiple peer rela-
tions problems needs to be examined. Most importantly, additional theory
and research is needed on the mechanisms by which these processes oper-
ate.
Because mechanisms underlying the link between the various types of
children's peer relations problems and school maladjustment have not
been studied directly, we reviewed the literature on social self cognition
factors that mediate the effects of interpersonal stress on adult adjustment.

The next section provides a brief review of this adult literature and our
conclusions about how these findings provided direction for the develop-
ment of our theory of social self cognitions as mediators of the effects of
children's peer relations problems on school maladjustment.
Social needs as a mediator of adult maladjustment

It has long been hypothesized that deficits in the formation of social rela-
tionships (Fromm-Reichmann, 1959; Sullivan, 1953) or the loss of social
relationships (Bowlby, 1980) lead to negative outcomes. These ideas have
been explored in the adult literature, with particular attention to loneliness
and depression as outcome measures. The results of these studies have
been mixed. Russell, Peplau, and Cutrona (1980) found that loneliness
correlated positively with time spent alone, whereas others found no sig-
nificant relation between number of friends reported and loneliness
(Jones, 1982; Stokes, 1985; Williams & Solano, 1983). In the past decade,
a great deal of research has been devoted to explaining individual differ-
ences in the extent to which the adequacy of adults' social relationships
predicts loneliness or depression. The research in each of these areas has
identified social needs as one of the mediating variables between actual
social relationships and these specific negative outcomes for adults.
The adult loneliness literature has suggested that two types of loneli-
ness exist and are related to different social needs (Weiss, 1973). Weiss
(1973) proposed that emotional loneliness occurs when an intimate at-
tachment is not present. He distinguished emotional loneliness from so-
cial isolation, which occurs when a network of social relationships is ab-
sent or inadequate (Weiss, 1973). Weiss' theory suggests that the presence
of specific kinds of relationships is necessary for everyone, and that lone-
liness can be predicted by simply measuring all of the relationships that
exist in a person's life. Later work has shown that the quality of relation-
ships is more important than the quantity in making this prediction (Rook,
1987; Shaver & Buhrmester, 1983). Several researchers have theorized
that the prediction of loneliness is dependent on the degree of discrepancy
between an individual's desired and actual social relationships (Peplau &
Perlman, 1982; Rook, 1988). From this perspective, loneliness is a func-
tion of the discrepancy between an individual's social needs and the indi-
vidual's perception of his or her actual social relationships.
The adult depression literature had taken a similar approach, consider-
ing the interactions among an individual's social needs, actual social rela-
tionships, and psychological adjustment. Beck (1983) proposed that per-

sonality traits could be used to predict which individuals would become
depressed when faced with certain kinds of life stressors. Specifically,he
hypothesized that sociotropic individuals (individuals who are dependent
on interpersonal relationships for safety, help, and gratification and highly
dependent on feedback from others) will be prone to depression when re-
jection by or separation from others is experienced, whereas autonomous
individuals (individuals who are dependent on meeting achievement-relat-
ed internal standards and goals and are less dependent on feedback form
others) will be relatively unaffected by these social stressors. In support of
this theory, several researchers have found that depression in sociotropic
adulsts is correlated with negative social events (Hammen, Ellicott, Gitlin,
& Jamison, 1989; Robins & Block, 1988; Robins, 1990). From this per-
spective, then, depression is a function of an interaction between an indi-
vidual's social needs (conceptualized as a personality trait) and negative
changes in his or her actual social resources.
In both the loneliness model and the depression model, the discrepancy
between social needs and perceived social relationships is thought to be a
cognitive mediator between social relationships and psychological adjust-
ment. Although this idea had been supported empirically for adults, it has
not yet been extended to the study of children's social relationships. The
adult psychiatric literature led us to consider the application of individual
differences in social needs as underlying the negative impact of children's
peer relations problems on school maladjustment.
The model we will propose differs from the adult models in several
ways. First, in the adult literature, the social needs construct is conceptual-
ized as a global personality variable (such as sociotropy or need for relat-
edness). In contrast, we are conceptualizing social needs as being domain-
specific. In the case of children's social needs, domain-specificity refers
to the type (e.g., group, dyadic), quality (e.g., conflict, support), and quan-
tity of peer relations. We theorize that a child's social needs within each
domain will be relatively independent of his or her social needs in other
domains. Second, the adult literature guided us only in terms of develop-
ing theories related to the specific outcomes of depression and loneliness
as a function of the discrepancy between social needs and actual social re-
sources. Although both of these affective problems are associated with
school performance and behavior, we were interested in developing a
model that would help to explain a broader range of school outcomes in-
cluding absenteeism, disruptive behavior, and aggression in school. Thus,
we reviewed the social psychological literature on the social self. Finally,
the adult literature presents a static model of how mental health may be af-
fected by interpersonal stress exposure and stable individual differences in
social needs. In contrast, our model is conceptualized from a developmen-
tal perspective, with the expectation that social needs will differ not only
across individuals, but also across developmental periods.
Given that the adult literature guided us to examine social needs as a
mediator of peer relations problems on school maladjustment, we began
the development of our theory by defining the construct, thinking about
the domains or features of social needs, and by examining any normative
data on social needs in children of different ages. We were also interested
in gender, cultural, and individual differences in reported social needs.
The next section reviews these points.
Characteristics of social needs

Definition of social needs
The study of children's social needs has not been directly examined in the
empirical literature on social development. The literature that comes clos-
est to examining this construct focuses in the study of children's social
goals (see also Ford, Berndt & Keefe, Erdley, Wentzel, this volume). A
"social goal" is defined within a social information-processing model as a
type of social cognition that may determine the choice of strategies used
to solve a social problem. Social goals have been operationalized in terms
of the proximal and immediate goals for specific social situations (e.g.,
Renshaw & Asher, 1983). Although this line of research has proven fruit-
ful, the present chapter defines social needs in a broader context than the
more behavioral or instrumental definitions used for the study of social
goals. Social needs are thought to reflect cognitive structures representa-
tive of the social self. These structures are thought to be chronically acces-
sible in memory and capable of influencing social goals and behavior in
specific situations (Sedikides & Skowronski, 1990, 1991; see also Hig-
gins, 1990).
In our model, a social need is defined as an individual's subjective
evaluation of the importance of and desire for various types, qualities, and
quantities of social and interpersonal relations. For example, a person for
whom having a best friend is viewed as very important would be said to
have a high level of need for that type of relationship. Likewise, a person
to whom being popular is relatively unimportant would be said to have a
low level of need for popularity. Thus, social needs may or may not be sit-
uationally specific and may influence, but do not refer to situationally

specific goals.
Domains of social needs

We use three broad domains to organize our discussion of social needs.
First, social needs may vary with regard to the type of social need; specif-
ically, whether the type of desired peer relation is a dyadic or group affili-
ation or a particular status. In other words, some children may desire a
best friend, others may want a group with which to hang out, and still oth-
ers may want to be popular. There are at least two aspects of the type of
peer relations that also are likely to vary across individuals. First, in addi-
tion to desiring a specific type of relationship, a child may target a. par-
itcular person to fulfill that need. For example, a boy may want to have a
girlfriend ( a certain type of relationship) and may want a particular girl to
be his girlfriend (a specific target). In addition, social needs may vary
with regard to specific characteristics of the target of the relationship,
such as gender. For example, a child may want to be friends with a group
of boys. Another child may desire to have a best friend who is very popu-
lar with the peer group. In these examples, the characteristics of the tar-
geted peer(s) are important components of the type of desired relation-
ship.
Second, social needs may vary with regard to the qualities of the peer
relationship or the functions provided by the relationship, such as admira-
tion, acceptance, avoidance of rejection, companionship, intimacy, aid,
safety, affection, enhancement of worth, trust, and resolution of conflict
(see Furman & Buhrmester, 1985; Weiss, 1974, for a discussion of rela-
tionship qualities). For example, one child may have a need for compan-
ionship and not a need for intimacy, whereas another child may report a
need for both qualities. In addition, the qualities or provisions of peer rela-
tionships desired by children may be experienced in varying levels. Chil-
dren may have a high need for trust, a moderate need for companionship,
and a low need for admiration.
Third, social needs may vary in terms of the quantity of relationships
that a child desires. One child may have an extremely high level of need
for a large quantity of best friends and be dissatisfied with less than five
'best friends.' Another child may also have a high need for best friend re-
lationships, but desire and be satisfied with having one best friend.
In addition to social needs varying in terms of their domains or fea-
tures, children may vary in terms oftheir flexibility or rigidity in meeting
particular social needs. For example, a child may be satisfied only with a
best friend relationship with a particular person (indicating low flexibili-
ty) and may ahve difficulty having this need fulfilled by this particular
person. Rigidity in social needs may prove to be frustrating and problem-
atic for children, because it may be more difficult for them to have their
social needs met. Other children may be more flexible in having their so-
cial needs met and experience less discomfort as a result of their social re-
lationships. In addition, social needs may vary in terms of their complexi-
ty. For example, one child may need a best friend, a boyfriend, and to be
popular in order to feel fulfilled, whereas another child may feel fulfilled
simply by having a best friend.
Perceptions of social needs may also vary in terms of their develop-
mental timing and appropriateness. For example, nonnormative social
needs such as a precocious need for a physically intimate relationship with
a member of the opposite sex may present problems for children.
Finally, social needs related to peers may compete at a particular point
in time or in a particular setting with other needs such as the need for
achievement. For example, a child who wants friends more than good
grades may choose to neglect his homework rather than jeopardize a
friendship by being unavailable after school to play.
Group differences in social needs

Developmental differences in social needs. The specific social needs of
children and adolescents as a function of development are not well under-
stood due to a paucity of research on this topic. The examination of devel-
opmental trends in social needs has been limited to cross-sectional work
with specific age groups and, for the most part, has examined self-reports
of social goals or the relative importance of different kinds of relation-
ships. Research on children's social networks suggests that children en-
gage in different types of relationships at different ages and that relation-
ships with different people are reported to be relatively more important
across development (Furman & Buhrmester, 1985). Specifically, students'
interest, dependency, and conformity to peers increases from childhood
through adolescence as their reliance on parents decreases (Berndt, 1979;
Costanzo, 1970; Steinberg & Silverberg, 1986). We might speculate, then
that the overall decline in achievement task values reported by Wigfield
and Eccles (1992) may reflect a developmental shift to a higher valuation
of tasks that meet social needs.
The relative importance of having a best friend increases across child-
hood and peaks in preadolescence (Furman & Buhrmester, 1985). For ex-
ample, most children (78%) in late elementary school have at least one
friend; however, only about one-half have a reciprocated best friend (Park-
er & Asher, 1993). In addition, the importance of intimacy within the best
friend relationship increases in middle childhood through early adoles-
cence (Berndt & Perry, 1986; Furman & Bierman, 1983). In our focus
groups, having a best friend was seen as extremely important by late ele-
mentary school-aged children: "Everybody has a best friend." "Most girls
want a best friend." When asked what kids would think about someone
who didn't want a best friend, these children responded, "If a girl didn't
want a best friend, they might pick on her," and "I've never had a kid [who
didn't want a best friend] in my grade." In contrast, adolescents saw hav-
ing a best friend as less important: "As long as [someone] has friends, it
doesn't matter." "It don't make no difference one way or the other."
In terms of the importance of broader peer group relations, belonging
to a crowd is extremely important to young teenagers and reaches its peak
at age 12 or 13 (Brown, Eicher, & Petrie, 1986). The importance attached
to crowd membership steadily declines across mid- to late adolescence
(Brown et al., 1986). Late elementary school-aged children seemed to
view the concept of "a group of freinds" as implying an organized, some-
what structured group. They spoke of informal "clubs" that children
"join," and several children indicated that groups of students had tests that
an individual must pass in order to join the group. The younger children
agreed that, although most students wanted to be part of a group, being
part of a group is not very important as long as a child has a friend. Ado-
lescents, however, reported that being part of a group of friends is most
important to them: "It's really important to fit in." They reported that chil-
dren who did not want to be part of a group would be thought of as "stuck
up" or "conceited." They talked about groups of friends in terms of "hang-
ing out" together, and about acting like the group members in order to
gain acceptance: "OK, this group likes me, I have to do everything they
do."
In one study, children reported that the importance of being popular
decreases from late childhood through late adolescence (Epstein, 1983a).
Interestingly, although children in our focus groups reported that the im-
portance of being popular decreased for themselves, they reported that
they thought being popular was always important to other students in their
grade. Children at all ages in the focus groups agreed that being popular
was important to some, but not all children (a 7th grade girl said, ". . .but
there are some people who just don't care about being popular"); however,
definitions of popularity were very different across different ages.

Younger children said that in order to be popular, a person would have to
"make a lot of friends," "get every answer right," "be really smart," and
"always act . . . like a kiss-up to the teacher." Older children had a more
complex definition of popularity, and discussed popularity in terms of so-
cial salience: "There's two kinds of popularity - one where everybody
likes you and one where everybody hates you."
Cross-sex friendships are almost nonexistent in elementary school;
however, they steadily increase across adolescence (Eder & Hallinan,
1978). Sharabany, Gershoni, and Hoffman (1981) reported that adoles-
cents' reports of intimacy in cross-sex friendships increase across ado-
lescence as well, perhaps replacing the importance of the larger peer
group or the same-sex best friend. In our focus groups, the importance
of having a boyfriend or girlfriend appeared to increase with age.
Younger children in the focus groups indicated that it would be accept-
able for a person either to want or to not want a boyfriend or girlfriend.
They talked about "act[ing] yourself" and "be[ing] nice" as ways to get
boyfriends or girlfriends. Older children, in contrast, said it was "very
important" to have a boyfriend or girlfriend, and that if someone did not
want a boyfriend or girlfriend, kids would think there was "something
wrong with her."
We were able to locate only one study that examined developmental
patterns of younger children's social goals. Schmidt, Ollendick, and
Stanowicz (1988) studied 6- to 13-year-old children, and found that older
children tended to be more flexible in adapting social strategies to their
assigned goals. Although these results do not indicate differences in social
goals per se, they suggest a cognitive flexibility that may affect the devel-
opment of social needs. A number of studies have examined adolescent
populations, with the goal of understanding changing needs that charac-
terize adolescence (Wentzel, 1993; Ford, 1982; Sewell, Farley, Manni, &
Hunt, 1982). Ford (1982) found no difference between ninth and twelfth
grade students' social goals, suggesting that for adolescents, social goals
remain consistent over time. Wentzel (1993) and Sewell et al. (1982) re-
ported significant differences in the content of social goals within adoles-
cent populations, but these differences were identified as a function of
gender and ethnicity rather than developmental level. In addition, the lim-
ited age range examined in their studies did not provide information for
comparison between the differences in the specific social goals of chil-
dren and those of adolescents. Clearly, more developmental research is
needed in this area.
Gender differences in social needs. Gender differences have also been

identified as an important factor in examining the definition and develop-
ment of children's social needs and goals. However, as was found in the
literature on developmental differences on social needs, the influence of
gender is not well understood. Research on the affiliation patterns of chil-
dren as a function of gender suggest that upon entrance to elementary
school, boys name more friends than do girls in the same class (Tuma &
Hallinan, 1979). Eder and Hallinan (1978) found that girls tend to have
more exclusive friendships than boys, and that boys report larger social
networks than girls (Clark & Ayers, 1988; Maccoby & Jacklin, 1974); al-
though no sex differences in the size of average friendship groups were re-
ported in adolescence (Hansell, 1981). Sherman (1984) reported that girls
evidence greater intimacy and sociability with same-sex friends than do
boys. In addition, Clark and Ayers (1988) reported that friendships be-
come increasingly reciprocal for girls as age increases. Boys report less
intimacy, caring, instrumental aid, and more difficulty resolving conflicts
with their best friend than do girls (Bukowski, Gauze, Hoza, & Newcomb,
1993; Parker & Asher, 1993). Males also report that popularity is more
important to them than it is to females (Epstein, 1983a). Most of these
studies do not address differences in social needs by gender; however, the
findings suggest that girls possess stronger social needs for intimacy than
do boys, whereas boys report greater need for popularity than do girls.
In our focus group discussions, gender differences emerged in discus-
sions of boyfriend/girlfriend relationships and popularity. Older girls in-
dicated that having sex was a necessary step to keep a boyfriend: "If you
don't do it with 'em they think you're boring." "[If you don't have sex]
they drop you right then." Boys, in contrast, did not identify sex as nec-
essary to keep a girlfriend. Older children also saw popularity as differ-
ent for boys and for girls. They reported that to be popular, girls "have
sex," "play sports," and "get their hair fixed," whereas boys "go to a dif-
ferent crowd of people and change your friends . . . get with a different
girl-
Research on social goals provides additional evidence for differential
goals as a function of gender. Wentzel (1993) found that early adolescent
girls reported higher academic and social responsibility goals than males.
In contrast, Sewell et al. (1982) reported that within an all-African-Amer-
ican adolescent population, males demonstrated greater achievement mo-
tivation than females. These mixed findings may be attributed to the use
of different measures in each study, as well as to possible interactions be-
tween gender and ethnicity. Overall, these findings provide important ad-
ditional evidence regarding the presence of multiple and possibly conflict-

ing goals in school in adolescence.
Ethnic or cultural differences in social needs. A final sociodemographic

variable examined in the context of social needs and goals is ethnicity or
culture. We were able to locate only a few studies that addressed the contri-
bution of ethnicity or culture to social needs. Wentzel (1993) reported that
early adolescent Caucasian children demonstrated higher social responsi-
bility goals than non-Caucasian children. Sewell et al. (1982) studied an
African-American sample of adolescents with regard to achievement goals,
but did not compare this group to a non-African-American population. Al-
though Wentzel's findings suggest that cultural factors affect social goals,
they do not explain the development of these cultural differences nor the
mechanisms by which ethnicity or culture influence children's and adoles-
cents' social needs.
Research on differential friendship patterns as a function of ethnicity
suggest that children tend to play or be friends with others from their same
ethnic group (Kupersmidt, DeRosier, & Patterson, 1995). In addition,
African-American youth report somewhat different qualities in their
friendships compared to Caucasian youth. For example, African-Ameri-
can males and females reported greater intimacy and support from peers
than did Caucasian males, and less than Caucasian females (DuBois &
Hirsch, 1990). DuBois and Hirsch (1990) also found that African-Ameri-
can peer networks included more neighborhood relationships than did
Caucasian networks, although the size of the in-school networks did not
differ. Clark and Ayers (1988) reported that African-American students
made more nonreciprocated friendship choices compared to Caucasian
students. Again, these findings are limited in their applicability to the
characterization of social needs in different ethnic groups, but the results
offer some indication that African-American and Caucasian students may
differ in their affiliation needs with Afncan-American youth possessing a
greater need for affiliation, support, and intimacy than Caucasian youth.
Individual differences in social needs

In order for social needs to be useful as an important mediator for explain-
ing individual differences in response to peer relationship problems, evi-
dence demonstrating individual differences in social needs would need to
be observed. Individual differences in distress about one's social experi-
ences could provide additional evidence in support of this hypothesis. We
located several studies that provide initial support for each of these as-
sumptions. For example, Crick and Ladd (1993) reported considerable
within-social status variation in social distress (e.g., 16% of popular chil-
dren and 44% of unpopular children reported high levels of social dis-
tress). In addition, Asendorpf (1993) has described three different types of
children who may exhibit high rates of solitary behavior. He suggested
that one type of child may prefer being alone to social activity and also
prefer to play with toys or do school work (Asendorpf, 1991; Coplan, Ru-
bin, Fox, Calkins, & Stewart, 1994). In contrast, another type of child may
want to interact with his or her peers, but does not do so. Their social
withdrawal is most obvious in novel settings where they fear negative
evaluation of others (Asendorpf, 1991). A third group may consist of chil-
dren who desire social interactions with others and do not try to isolate
themselves, yet they may be isolated by peers due to their incompetent so-
cial behavior (Coie & Kupersmidt, 1983; Dodge, 1983; Rubin & Mills,
1988). Thus, taken together, these studies suggest that there are individual
differences in levels of social needs and in the importance of different
types of social interactions across individuals.
In addition, several studies have reported differences in children's so-
cial goals as a function of having different kinds of peer relations prob-
lems or behavior problems that are associated with peer relations prob-
lems. In conflict situations, unpopular or rejected children are more
focussed on instrumental rather than relational outcomes (Crick & Ladd,
1990) and rank-order positive, social goals lower (Renshaw & Asher,
1983) than do more popular children. Likewise, very aggressive youth en-
dorse goals that are hostile in nature (Slaby & Guerra, 1988) and place
high value on dominance or control of peer victims (Boldizar, Perry, &
Perry, 1989; Lochman, Wayland, & White, 1993) as compared to less ag-
gressive youth. Taken together, these findings are consistent with theoreti-
cal work that suggests that less socially competent children have problems
prioritizing goals and coordinating multiple goals as compared with so-
cially competent youth, particularly in conflict situations (Dodge, Asher,
&Park, 1989).
Social self-discrepancy theory, problematic peer relations,

and school maladjustment
We conducted a review of the social psychological literature on self-cog-
nitions to attemtpt to explain the mechanism by which the discrepancy be-
tween adults' social needs and their social resources produced loneliness
or depression. Self-discrepancy theory, developed by Higgins and his col-

leagues (Higgins, 1987; Higgins, Bond, Klein, & Strauman, 1986; Strau-
man & Higgins, 1987), provides an excellent framework for understand-
ing social needs as a mediator of the effects of problematic peer relations
on school maladjustment. Self-discrepancy theory distinguishes between
domains of the self and standpoints on the self. Two domains of the self
were relevant to explaining the discrepancy of interest, namely, the actual
self (representation of attributes actually possessed) and the ideal self
(representation of attributes that might ideally be possessed). In addition,
the theory emphasizes two standpoints on the self: own (representations
that stem from a personal point of view) or other (representations that
stem from the point of view of a significant other.)
There is an important difference between the original conceptualiza-
tion of self-discrepancy theory, as described in the above paragraph, and
our adaptation of the theory for the purposes of our research. The original
proposal addressed self-discrepancies within the totality of self- represen-
tations, whereas we are concerned with self-discrepancies within the rep-
resentations that are relevant to the social self. Specifically, we define the
actual social self as children's actual social resources (e.g., the type, quali-
ty, and quantity of peer relations). We define the ideal social self as chil-
dren's subjective accounts of their ideal social resources or their social
needs (e.g., the desired type, quality, and quantity of peer relations). Con-
sistent with the original theory, these subjective accounts of the actual so-
cial self can represent the standpoint of either the children themselves or
significant others (e.g., peers, parents, teachers; see Hartup, 1989, and
Higgins, 1991, for discussions of the important socializing roles of peers,
parents, and teachers; see also Harter, this volume).
Studies with adults have found that discrepancies between the actual
self and the ideal self are strongly associated with dejection-related emo-
tions such as feeling sad, disappointed, or discouraged (Higgins et al.,
1986; Higgins, Klein, & Strauman, 1985; Strauman & Higgins, 1987). In
our case, then, a discrepancy between one's ideal social self (social needs)
and one's actual social self would be expected to be associated with dejec-
tion-related affect. Although this discrepancy has not been studied directly
among children, several studies have examined children's subjective expe-
riences or reports of different types of desired peer relations. For example,
the lack of best friendships, low peer acceptance, and poor quality friend-
ships were all associated with loneliness when examined in the context of
hierarchical regression models (Parker & Asher, 1993). This theory would
speculate, then, that loneliness could be predicted from the discrepancy
between children's desire for better peer relations and their lack of desired
peer relations. Likewise, adolescents without reciprocated or stable
friends report that popularity is more important to them than do adoles-
cents with reciprocated or stable friends (Epstein, 1983a). In addition, in
middle school, submissive-rejected students report being lonelier than av-
erage children (Parkhurst & Asher, 1992).
The original self-discrepancy theory proposes that affective reactions,
such as dejection-related affect, may lead to motivated thoughts or behav-
iors that are likely to alleviate the impact of negative affect. Motivated
thinking or behaving include either changing cognition (e.g., changing the
content of the actual or ideal social self) and/or changing behavior (i.e.,
behaving in ways that are likely to reduce the gap of the social self-dis-
crepancy) (Aronson, 1969).
We should note here, however, that we are not arguing that the affect
associated with social self-discrepancies is the only motivator of behavior.
In addition, we suggest that the lack of discrepancy between social needs
(the ideal social self) and social resources (the actual social self) may also
be associated with an affective reaction that may also serve as a strong
motivatior of behavior in children. For example, children who want to be
well-liked and are popular may be highly motivated to attend school be-
cause it is a rewarding and satisfying environment. In fact, in one of our
focus groups, a 7th grade girl reported, "If your are popular, you are going
to want to come to school to keep up your reputation." The lack of dis-
crepancy does not guarantee positive school outcomes as suggested by an-
other 7th grade girl who responded to a question about whether being
popular and wanting to stay popular could affect a student's schoolwork
by saying, "Yes, because you're so busy concentrating on being popular,
you just don't even think about your schoolwork."
Children who experience dejection-related affect associated with and
idealractual social self-discrepancy may be motivated to reduce the dis-
crepancy by changing their cognitions or their behavior. One way that
children may change their cognitions is by changing their ideal social self.
For example, a child for whom peer acceptance is important (ideal social
self), but who is rejected by her peers (actual social self) may change her
ideal social self. She may, over time, develop a new ideal social self that
elevates the desire for a same-sex best friend above the desire to be ac-
cepted by her peers. If she does acquire a same-sex best friend, the dis-
crepancy that produced the negative affect will be reduced.
The other method for reducing a social self-discrepancy, behavior
change, is perhaps more relevant to the goals of the present chapter. Fig-
ure 4.1 depicts a hypothetical example in which the negative affect associ-
ated with the ideal:actual social self-discrepancy motivates behavioral
change relevant to school maladjustment. In Figure 4.1, the discrepancy is
reduced through avoidance of the situational (i.e., school) cues that acti-
vate this discrepancy. This pattern would be consistent with literature on
depression and school adjustment that suggests a significant correlation
between depressed affect and school absenteeism (e.g., Young, Brasic, &
Kearney, 1993; Kisnadwala, 1990). In this example, then depressed affect
is associated with school absenteeism as the student attempts to avoid the
cause of the depressed feelings.
Several examples from our focus groups illustrate the association be-
tween ideahactual social self-discrepancies and dejection-related affect. A
third grade boy said "If [a boy]'s disliked he wouldn't want to go to school
'cause the kids don't like him." Likewise, a ninth grade girl said about
kids who are disliked, "You'll see 'em one week, then they'll talk junk to
somebody and seven girls'll jump 'em and they won't be back." A 7th
grade female student reported that being disliked and wanting to be liked
could produce dejection-related emotions and affect going to school in
that "it would, it might make them want to drop out, if they are disliked -
this girl in my sister's school, people didn't like her and talked about her
and she committed suicide and her mother committed suicide because
they had so much stress and stuff and her brother, he went to her school,
they were still talking about her family and stuff and he still wanted to
commit suicide." In this example, two behavioral strategies for reducing
IDEAL SOCIAL SELF:

It is very important to me
to be accepted by kids in
my classroom.
SOCIAL SELF AFFECTIVE REACTION: BEHAVIOR CHANGE:

I am lonely and depressed I often stay home
DISCREPANCY because the kids are mean from school.
tome.
\
ACTUAL SOCIAL SELF:

Many kids in my class
don't like me. They make
fun of me every day.
Figure 4.1. An example of a school-related behavior change resulting

from an ideahactual social self discrepancy.
an ideahactual social self-discrepancy were implied, namely, avoiding

school and committing suicide. This example provides a clear illustration
of how the ideahactual social self-discrepancy can impact school malad-
justment. These findings are consistent with the hypothesis suggested by
Coie (1990) that a child who wants to be liked by members of his or her
peer group, but who is rejected by them, may avoid peers to avoid further
rejection and to decrease exposure to negative behaviors directed at them
by peers.
In addition to the ideal and actual domains of the self, self-discrepancy
theory introduces another self-cognition that is relevant to broadening the
scope of our theory. This third domain is termed the ought self, and is de-
fined as a representation of attributes that ought to be possessed. In our
case, the ought domain would be defined as the norms or standards for so-
cial development (e.g., the type, quantity, and quality of peer relations that
a child of a particular age "should" possess). Higgins (1987) hypothesized
that a discrepancy between the actual self and the ought self would be as-
sociated with agitation-related problems such as feeling worried, nervous,
or tense. Several adult studies have reported a relation between the actu-
al:ought discrepancy and anxiety-related symptoms (Higgins et al., 1986;
Higgins et al., 1985; Strauman & Higgins, 1987). In the social domain, a
discrepancy between one's social standards and one's actual social ability
or resources might be expected to produce social anxiety. In this case, for
example, socially rejected children who think they should be well-liked
(as opposed to ideally wanting or needing to be well-liked by peers) would
be expected to be socially anxious if they perceived themselves to be dis-
liked by peers.
Social anxiety could function as a motivator for either a change in cog-
nitions or a change in behavior that would reduce the ideal-ought social
self-discrepancy. A child with a social standard that isn't met might be
flexible and change this standard. This change may result in a reduced dis-
crepancy which may, in turn, reduce anxiety-related affect. In the school
setting, social anxiety might lead to a refusal to speak or perform in front
of others in class and a timidness in social situations that form the fabric
of everyday life in school. Figure 4.2 shows a hypothetical example of one
way in which the ought:actual social self-discrepancy could lead to behav-
ior change with implications for school maladjustment. Figure 4.2 depicts
one component of this pathway, in that the discrepancy is reduced by a
change in the actual social self that is accomplished by a behavior change
of not participating in class.
One example from our focus group discussions illustrates the
OUGHT SOCIAL SELF:

1 should be welMiked and
respected by my
classmates.
AFFECTIVE REACTION:
BEHAVIOR CHANGE:
SOCIAL SELF I am worried when I go to
I never answer questions
DISCREPANCY school that the kids will
or read out loud in class.
call me names. I am
\ afraid of being embarrasse<
or humiliated in class.
ACTUAL SOCIAL SELF:
The kids at school don't
like me and make fun of me.
Figure 4.2. An example of a school-related behaivor change resulting

from an ought:actual social self discrepancy.
actuahought social self-discrepancy was "She may not try to answer a

question that she really knows, because she's scared she'll get it wrong,
and she don't want to be really, really disliked." Although there are no di-
rect studies on this actuahought social self-discrepancy hypothesis for
children, some studies suggest support for this hypothesis. For example,
Durlak (1992) found that anxiety-related affect was related to school re-
fusal and school phobia, and that children with internalizing disorders
demonstrated heightened test and performance anxiety. Also, submissive-
rejected middle school students report more concern about the possibility
of being rejected or humiliated after lunch and between classes than do
average students (Parkhurst & Asher, 1992). Rejected children reported
more anxiety of the form of greater concern about peer evaluations than
did more accepted peers (La Greca, Dandes, Wick, Shaw, & Stone, 1988).
A final mechanism is based upon the work of Ogilvie (1987) who in-
troduced the concept of the undesired self. This notion is consistent with
Sullivan's (1953) ideas about the "good" me, the "bad" me, and the "not"
me. Applying Sullivan's theory, the ideal self would be derived from the
good me, whereas the undesired self would have images of the bad me and
the not me. Ogilvie argues that the "not me" is the most dangerous and
disowned image of the self, because these images are derived from memo-
ries of dreaded experiences, embarrassing situations, fearsome events, and
unwanted emotions. He argues that a discrepancy between the ideal self
and the not me or the undesired self may produce a strong negative affec-
tive reaction and perhaps would be even more motivating than the nega-
tive affect produced by other types of self-discrepancies.
A child who has developed an undesired social self, based on previous

negative experiences at school, then, would have a strong motivation to re-
duce this discrepancy, either through changes in cognition or behavior.
Figure 4.3 depicts a hypothetical example in which behavior change is
motivated by negative affect associated with an undesired:ideal social
self-discrepancy, with implications for school maladjustment. In figure
4.3, the discrepancy is reduced through avoidance of the undesired social
self. This avoidance is accomplished by the behavior change of affiliating
with an antisocial peer group rather than be friendless.
This modification to self-discrepancy theory provides an expanded
motivational framework for explaining observations and theories in the
peer relations literature, such as the "shopping" metaphor introduced by
Patterson, Littman, & Bricker (1967) to describe the developmental
process involved in the initiation and maintenance of friendships among
antisocial peers. Children and adolescents highly value having a group of
friends or being part of a peer cluster (Oetting & Beauvais, 1987). Howev-
er, there are some children who are disliked, have poor social skills, and
have difficulty developing friendships with peers. These individual differ-
ences provide the substrate or underlying conditions that may affect the
likelihood that an individual child will associate with others who have
problems with school, authority, and crime. Students who get into trouble
in school tend to affiliate with one another as well as to get into trouble to-
gether and have a modest negative influence upon one another (Berndt,
IDEAL SOCIAL SELF:

It is very important to me
to be friends with the
popular kids in my grade.
SOCIAL SELF AFFECTIVE REACTION: BEHAVIOR CHANGE:

DISCREPANCY I am afraid of being alone I'll hang out with anyone
and having no friends. who will be my friend. I'll
hang out with the kids
\ who get in trouble in school.
UNDESIRED SOCIAL SELF:

I've tried to be friends with the
popular kids, but they ignore
me. Lots of times no one will
play with me or talk to me.
Figure 4.3. An example of a school-related behavior change resulting

from an ideal:undesired social self discrepancy.
1992). In fact, a few longitudinal studies have demonstrated increases

over time in friends' similarity on antisocial behaviors and attitudes (Ep-
stein, 1983b; Kandel 1978a). The observed homophily among low status,
low achieving, and antisocial students may be explainable as a reaction to
the fear of being socially isolated and lack of access in establishing rela-
tionships with more desirable partners.
Self-discrepancy theory does not provide us with a means for explain-
ing why social influence processes occur. We believe that the tenets of so-
cial learning theory as well as recent theoretical work by Berndt (1992)
concerning mutual influence processes in friendships provide a better ex-
planatory base for a discussion of influence processes. However, self-dis-
crepancy theory provides a framework for understanding individual differ-
ences in the motivation behind the initiation of relationships with
problematic peers, as well as individual differences in the susceptibility to
peer influences on school misbehavior, school failure, and school dropout.
Conclusions
The model that we have proposed in this chapter represents an attempt to
integrate previous research on self-cognitions with the peer relations liter-
ature, with an emphasis on school maladjustment as an outcome. We have
attempted to go beyond description of the relations among different kinds
of peer relations problems and school maladjustment and to focus on a
possible mechanism that may underlie the observed relation between so-
cial and school functioning. We have proposed one broad framework with-
in which these mechanisms may be understood and tested. In this chapter,
we have argued that discrepancies among various aspects of the social self
can be powerful elicitors of negative affect which, in turn, may motivate
behavior.
In the present framework, our goal has been to focus on peer problems
and their relation to school problems; however, self-discrepancy theory
may also be applicable to explain the relation between social success and
school competence. This approach is not incompatible with other research
on motivation and social behavior. The social needs:social resources dis-
crepancy undoubtedly operates within a network of other motivators to
produce an affective outcome.
The ideas contained in this chapter outline a clear agenda for future
sesearch as described below. The peer relations field has been richly de-
veloped over the past two decades, and has begun to describe the topogra-
phy of social relations across different developmental periods. However,
cognitions about social relationships across different developmental levels

are less well known. We have proposed that one type of cognition, chil-
dren's desire for various types, qualities, and quantities of peer relations
may play an important role in the association between peer and school
problems and warrants systematic investigation. We have also introduced
two additonal social self-cognitions, including the ought and undesired
selves, that have not been previously investigated in children and that also
have potential explanatory power as mechanisms that may mediate the re-
lation between peer and school problems. In addition to the study of indi-
vidual differences in social as well as other needs, developmental, ethnic,
and gender differences need to be examined. The use of a framework such
as this will allow the field to move beyond the description of social rela-
tionships and their impact on youth to the development of explanatory
models of the processes involved.
Acknowlegments
This research was supported in part by a William T. Grant Faculty Schol-
ars Award and a University of North Carolina at Chapel Hill University
Research Council Grant to the first author. The authors would like to
thank the students and staff of the Durham Public School System for their
help in conducting the study reported in this chapter.
References
Aronson, E. (1969). The theory of cognitive dissonance: A current perspective. In
L. Berkowitz (Ed.), Advances in experimental social psychology (vol. 4, pp.
1-34). San Diego, CA: Academic Press.
Armacost, R. L. (1989). Perceptions of stressors by high school students. Journal
of Adolescent Research, 4, 443-461.
Asarnow, J. R. (1988). Peer status and social competence in child psychiatric in-
patients: A comparison of children with depressive, externalizing, and concur-
rent depressive and externalizing disorders. Journal of Abnormal Child Psy-
chology 16, 151-162.
Asendorpf, J. B. (1991). Development of inhibited children's coping with unfa-
miliarity. Child Development, 62, 1460-1474.
Asendorpf, J. B. (1993). Abnormal shyness in children. Journal of Child Psychol-
ogy, Psychiatry, and Allied Disciplines, 34, 1069-1081.
Beck, A. T. (1983). Cognitive therapy of depression: New perspectives. In P. J.
Clayton & J. E. Barrett (Eds.), Treatment of depression: Old controversies and
new approaches (pp. 265-290). New York: Raven Press.

Berndt, T. J. (1989). Obtaining support from friends in childhood and adoles-
cence. In D. Belle (Ed.), Children s social networks and social supports (pp.
308-331). New York: Wiley.
Berndt, T. J. (1992). Friendship and friends' influence in adolescence. Current Di-
rections in Psychological Science, 1, 156—159.
Berndt, T. J., Laychak, A. E., & Park, K. (1990). Friends' influence on adoles-
cents' academic achievement motivation: An experimental study. Journal of
Educational Psychology, 82, 664-670.
Berndt, T. J., & Perry, T. (1986). Children's perceptions of friendships as support-
ive relationships. Developmental Psychology, 22, 640-648.
Boldizar, J. P., Perry, D. G., & Perry, L. C. (1989). Outcome values and aggres-
sion. Child Development, 60, 571-579.
Bowlby, J. (1980). Attachment and loss. Vol. 3: Loss. New York: Basic Books.
Brown, B. B., Eicher, S. A., & Petrie, S. (1986). The importance of peer group
("crowd") affiliation in adolescence. Journal of Adolescence, 9, 73-96.
Bukowski, W. M., & Hoza, B. (1989). Popularity and friendship: Issues in theory,
measurement, and outcome. In T. J. Berndt & G. W. Ladd (Eds.), Peer relation-
ships in child development (pp. 15^5). New York: Wiley.
Bukowski, W. M., Gauze, C, Hoza, B., & Newcomb, A. F. (1993). Differences
and consistency between same-sex and other-sex peer relationships during ear-
ly adolescence. Developmental Psychology, 29, 255-263.
Cairns, R. B., Cairns, B. D., Neckerman, H. J., Gest, S. D., & Gariepy, J. L.
(1988). Social networks and aggressive behavior: Peer support or peer rejec-
tion? Developmental Psychology, 24, 815-823.
Clark, M. L., & Ayers, M. (1988). The role of reciprocity and proximity in junior
high school friendships. Journal of Youth and Adolescence, 17, 403-411.
Coie, J. D. (1990). Toward a theory of peer rejection. In S. R. Asher & J. D. Coie
(Eds.), Peer rejection in childhood (pp. 365-401). Cambridge, England: Cam-
Coie, J. D., Dodge, K. A., & Coppotelli, H. (1982). Dimensions and types of so-
cial status: A cross-age perspective. Developmental Psychology, 18, 557-570.
Coie, J. D., & Kupersmidt, J. B. (1983). A behavioral analysis of emerging social
status in boys' groups. Child Development, 54, 1400-1416.
Coie, J. D., Lochman, J. E., Terry, R., & Hyman, C. (1992). Predicting early ado-
lescent disorder from childhood aggression and peer rejection. Journal of Con-
sulting and Clinical Psychology, 60, 783-792.
Colton, M. E., & Gore, S. (Eds.) (1991). Adolescent stress: Causes and conse-
quences. New York: Aldine de Gruyter.
Coplan, R. J., Rubin, K. H., Fox, N. A., Calkins, S. D., & Stewart, S. L. (1994). Be-
ing alone, playing alone, and acting alone: distinguishing among reticence and
passive and active solitude in young children. Child Development, 65, 129-137.
Costanzo, RR. (1970). Conformity development as a function of self-blame. Jour-

nal of Personality and Social Psychology, 14, 366-374.
Crick, N. R., & Grotperer, J. (1994). Children's treatment by peers: Victims of re-
lational and overt aggression. Unpublished manuscript. University of Illinois at
Urbana-Champaign.
Crick, N. R., & Ladd, G. W. (1990). Children's perceptions of the outcomes of so-
cial strategies: Do the ends justify being mean? Developmental Psychology, 26,
612-620.
Crick, N. R., & Ladd, G. W. (1993). Children's perceptions of their peer experi-
ences: Attributions, loneliness, social anxiety, and social avoidance. Develop-
mental Psychology, 29, 244-254.
DeRosier, M. E., Kupersmidt, J. B., & Patterson, C. J. (1994). Children's academ-
ic and behavioral adjustment as a function of the chronicity and proximity of
peer rejection. Child Development, 65, 1799-1813.
Dishion, T. J., & Loeber, R. (1985). Adolescent marijuana and alcohol use: The
role of parents and peers revisited. American Journal of Drug and Alcohol
Abuse, 11, 11-25.
Dodge, K. A. (1983). Behavioral antecedents of peer social status. Child Develop-
ment, 54, 1386-1399.
Dodge, K. A., Asher, S., & Parkhurst, J. T. (1989). Social life as a goal coordina-
tion task. In C. Ames & R. Ames (Eds.), Research on motivations in education
(vol. 3). New York: Academic Press.
Dodge, K. A., Coie, J. D., & Brakke, N. P. (1982). Behavior patterns of socially re-
jected and neglected preadolescents: The roles of social approach and aggres-
sion. Journal of Abnormal Child Psychology, 10, 389-410.
DuBois, D. L., & Hirsch, B. J. (1990). School and neighborhood friendship pat-
terns of blacks and whites in early adolescence. Child Development, 61,
524-536.
Durlak, J. A. (1992). School problems of children. In C. E. Walker & M. C.
Roberts (Eds.), Handbook of clinical child psychology (2nd ed.). Wiley series
on personality processes (pp. 497-510). New York: Wiley.
Eder, D., & Hallinan, M. T. (1978). Sex differences in children's friendships.
American Sociological Review, 43, 237—250.
Elliot, D. S., Huizinga, D., & Ageton, S. S. (1985). Explaining delinquency and
drug use. Beverly Hills, CA: Sage Publications.
Epstein, J. L. (1983a). Examining theories of adolescent friendships in J. L. Ep-
stein & N. Karweit (Eds.), Friends in school: Patterns of selection and influ-
ence in secondary schools (pp. 39-61). New York: Academic Press.
Epstein, J. L. (1983b). The influence of friends on achievement and affective out-
comes. In J. L. Epstein & N. Karweit (Eds.), Friends in school: Patterns of se-
lection and influence in secondary schools (pp. 177-200). New York: Academ-
ic Press.
Ford, M. E. (1982). Social cognition and social competence in adolescence. De-

velopmental Psychology, 18, 323—340.
French, D. C, & Waas, G. A. (1985). Behavior problems of peer-neglected and
peer-rejected elementary-age children: Parent and teacher perspectives. Child
Fromm-Reichmann, F. (1959). Loneliness. Psychiatry, 22, 1-15.
Furman, W., & Bierman, K. L. (1983). Developmental changes in young chil-
dren's concepts of friendship. Child Development, 54, 549-556.
Furman, W, & Buhrmester, D. (1985). Children's perceptions of the personal rela-
tionships in their social networks. Developmental Psychology, 21, 1016-1024.
Goodyer, I. M, Wright, C, & Altham, P. M. E. (1989). Recent freindships in anx-
ious and depressed school age children. Psychological Medicine, 19, 165-174.
Hammen, C, Ellicott, A., Gitlin, M., & Jamison, K. (1989). Sociotropy/autonomy
and vulnerability to specific life events in patients with unipolar depression and
bipolar disorders. Journal of Abnormal Psychology, 98, 154-160.
Hansell, S. (1981). Ego development and peer friendship networks. Sociology of
Education, 54, 51-63.
Hartup, W. W. (1989). Social relationships and their developmental significance.
American Psychologist, 44, 120-126.
Higgins, E. T. (1990). Personality, social psychology, and person-situation rela-
tions: Standards and knowledge activation as a common language. In L. A. Per-
vin (Ed.), Handbook ofpersonality: Theory and research (pp. 301—338). New
York, NY: Guilford Press.
Higgins, E. T. (1991). Development of self-regulatory and self-evaluative process-
es: Costs, benefits, and tradeoffs. In M. R. Gunnar & L. A. Sroufe (Eds.), Self
processes and development: The Minnesota symposia on child psychiatry, (vol.
23), Hillsdale, NJ: Erlbaum.
Higgins, E. T., Bond, R. N., Klein, R., & Strauman, T. (1986). Self-discrepancies
and emotional vulnerability: How magnitude, accessibility, and type of discrep-
ancy influence affect. Journal of Personality and Social Psychology, 51, 5-15.
Higgins, E. T., Klein, R., & Strauman, T. (1985). Self-concept discrepancy theory:
A psychological model for distinguishing among different aspects of depres-
sion and anxiety. Social cognition, 3, 51-76.
Johnson, J. H. (1988). Life events as stressors in childhood and adolescence.
Newbury Park: Sage Publications.
Jones, R. T., & Fiore, M. (1991, Nov). Conceptualization of self-generated stres-
sors: Alternatives, responses, and mediators. Paper presented at the 25th annu-
al meeting of the Association for the Advancement of Behavior Therapy, New
York, New York.
Jones, W. H. (1982). Loneliness and social behavior. In L. A. Peplau & D. Perlman
(Eds.), Loneliness: A sourcebook of current theory, research, and therapy. New
York: Wiley.
Kandel, D. B. (1978a). Homophily, selection, and socialization in adolescent

friendships. American Journal of Sociology, 84, 427-436.
Kandel, D. B. (1978b). Similarity in real-life adolescent friendship pairs. Journal
of Personality and social Psychology, 36, 306-312.
Kearney, C. A. (1993). Depression and school refusal behavior: A review with
comments on classification and treatment. Journal of School Psychology, 31,
267-279.
Kindermann, T. A. (1993). Natural peer groups as contexts for individual develop-
ment: The case of children's motivation in school. Developmental Psychology,
29, 970-977.
Kohlberg, L., LaCrosse, J., & Ricks, D. (1972). The predictability of adult mental
health from childhood behavior. In B. Wolman (Ed.), Manual of child psy-
chopathology (pp. 1217-1284). New York: McGraw-Hill.
Kupersmidt, J. B., Burchinal, M, & Patterson, C. J. (1995). Developmental pat-
terns of childhood peer relations as predictors of externalizing behavior prob-
lems. Development and Psychopathology, 7, 825-843.
Kupersimdt, J. B., & Coie, J. D. (1990). Preadolescent peer status, aggression, and
school adjustment as predictors of externalizing problems in adolescence.
Kupersmidt, J. B., Coie, J. D., & Dodge, K. A. (1990). the role of poor peer rela-
tionships in the development of disorder. In S. R. Asher & J. D. Coie (Eds.),
Peer rejection in childhood. Cambridge studies in social and emotional devel-
opment (pp. 274-305). New York: Cambridge University Press.
Kupersmidt, J. B., DeRosier, M. E., & Patterson, C. P. (1995). Similarity as the ba-
sis for children's friendships: The roles of sociometric status, aggressive and
withdrawn behavior, academic achievement, and demographic characteristics.
Journal of Social and Personal relationships, 12, 439-452.
Kupersmidt, J. B., Griesler, P., & Patterson, C. P. (1994). Sociometric status, ag-
gression, and affiliation patterns of peers. Unpublished manuscript. University
of North Carolina at Chapel Hill.
Kupersmidt, J. B., & Patterson, C. P. (1991). Childhood peer rejection, aggression,
withdrawal, and perceived competence as predictors of self-reported behavior
problems in preadolescence. Journal of Abnormal Child Psychology, 19,
427-449.
La Greca, A. M, Dandes, S. K., Wick, P., Shaw, K., & Stone, W. L. (1988). Devel-
opment of the Social Anxiety Scale for Children: Reliability and concurrent va-
lidity. Journal of Clinical Child Psychology, 17, 84-91.
Lochman, J. E., Wayland, K. K., & White, K. J. (1993). Social goals: Relationship
to adolescent adjustment and to social problem solving. Journal of Abnormal
Child Psychology, 21, 135-151.
Maccoby, E., & Jacklin, C. (1974). The psychology of sex differences. Stanford,
CA: Stanford University Press.
O'Brien, S. E, & Bierman, K. L. (1988). Conceptions and perceived influence of
peer groups: Interviews with preadolescents and adolescents. Child Develop-

ment, 59, 1360-1365.
Oetting, E. R., & Beauvais, F. (1987). Peer cluster theory, socialization character-
istics, and adolescent drug use: A path analysis. Journal of Counseling Psy-
chology, 34, 205-213.
Ogilvie, D. M. (1987). The undesired self: A neglected variable in personality re-
search. Journal of Personality and Social Psychology, 52, 379-385.
Ollendick, T. H., Weist, M. D., Borden, M. C, & Green, R. W. (1992). Sociomet-
ric status and academic, behavioral, and psychological asjustment: A five-year
longitudinal study. Journal of Consulting and Clinical Psychology, 60, 80—87.
Parker, J. G., & Asher, S. R. (1987). Peer relations and later personal adjustment:
Are low-accepted children at risk? Psychological Bulletin, 102, 357-389.
Parker, J. G., & Asher, S. R. (1993). Friendship and friendship quality in middle
childhood: Links with peer group acceptance and feelings of loneliness and so-
cial dissatisfaction. Developmental Psychology, 29, 611-621.
Parkhurst, J. T., & Asher, S. R. (1992). Peer rejection in middle school: Subgroup
differences in behavior, loneliness, and interpersonal concerns. Developmental
Patterson, G. R., & Dishion, T. J. (1985). Contributions of families and peers to
delinquency. Criminology, 23, 63-77.
Patterson, G. R., Littman, R. A., & Bricker, W. (1967). Assertive behavior in chil-
dren: A step toward a theory of aggression. Monographs of the society for re-
search in child development, (vol. 32, No. 5, Serial No. 113). Chicago, IL:
University of Chicago Press.
Peplau, L. A., & Perlman, D. (1982). Loneliness: A sourcebook of current theory,
research, and therapy. New York: Wiley.
Renshaw, P. D. & Asher, S. R. (1983). Children's goals and strategies for social in-
teraction. Merrill-Palmer Quarterly, 29, 353-374.
Robins, C. J. (1990). Congruence of personality and life events in depression.
Journal of Abnormal Psychology, 99, 393-397.
Robins, C. J., & Block, P. (1988). Personal vulnerability, life events, and depres-
sive symptoms: A test of a specific interactional model. Journal of Personality
and Social Psychology, 54, 847-852.
Rook, K. S. (1984). The negative side of social interaction: Impact on psycho-
logical well-being. Journal of Personality and Social Psychology, 46, 1097-
1108.
Rook, K. S. (1987). Social support versus companionship: Effects on life stress,
loneliness, and evaluations by others. Journal of Personality and Social Psy-
chology, 52, 1132-1147.
Rook, K. S. (1988). Towards a more differentiated view of loneliness. In S. W.
Duck (Ed.), Handbook of personal relationship: Theory, research and interven-
tions. Chichester: Wiley.
Rubin, K. H., Fein, G. G., & Vandenberg, B. (1983). Play. In E. M. Hetherington
(Ed.), Handbook of child psychology. Vol 4: Socialization, personality, and so-

cial development (pp. 693-774). New York: Wiley.
Rubin, K. H., & Mills, R. S. L. (1988). The many faces of social isolation in child-
hood. Journal of Consulting and Clinical Psychology, 6, 916-924.
Russell, D., Peplau, L. A., & Cutrona, C. E. (1980). The revised UCLA Loneliness
Scale: Concurrent and discriminant validity evidence. Journal of Personality
and Social Psychology 39, 472-480.
Schmidt, C. R., Ollendick, T. H., & Stanowicz, L. B. (1988). Developmental
changes in the influence of assigned goals on cooperation and competition. De-
velopmental Psychology 24, 574-579.
Sedikides, C, & Skowronski, J. J. (1990). Toward reconciling personality and so-
cial psychology: A construct accessibility approach. Journal of Social Behavior
and Personality, 5, 531-546.
Sedikides, C, & Skowronski, J. J. (1991). The law of cognitive structure activa-
tion. Psychological Inquiry, 2, 169-184.
Sewell, T. E., Farley, F. H., Manni, J. L., & Hunt, P. (1982). Motivation, social re-
inforcement, and intelligence as predictors of academic achievement in Black
adolescents. Adolescence, 17, 647-656.
Sharabany, R., Gershoni, R., & Hoffman, J. E. (1981). Girlfriend, boyfriend: Age
and sex differences in intimate friendship. Developmental Psychology, 17,
800-808.
Shaver, P., & Buhrmester, D. (1983). Loneliness, sex-role orientation, and group
life: A social needs perspective. In P. B. Paulus (Ed.), Basic group processes.
New York: Springer-Verlag.
Sherman, L. W. (1984). Social distance perceptions of elementary school children
in age-heterogeneous and homogeneous classroom settings. Perceptual and
Motor Skills, 58, 395-409.
Slaby, R. G., & Guerra, N. G. (1988). Cognitive mediators of aggression in ado-
lescent offenders: 1, Assessment. Developmental Psychology, 23, 580-588.
Steinberg, L., & Silverberg, S. B. (1986). The vicissitudes of autonomy in early
Stokes, J. P. (1985). The relation of social network and individual difference vari-
ables to loneliness. Journal ofPersonality and Social Psychology, 48, 981-990.
Strauman, T. J., & Higgins, E. T. (1987). Automatic activation of self-discrepan-
cies and emotional syndromes: When cognitive structures influence affect.
Journal of Personality and Social Psychology, 53, 1004-1014.
Sullivan, H. S. (1953). The interpersonal theory of psychiatry. New York: Norton.
Tesser, A., Campbell, J., & Smith, M. (1984). Friendship choice and performance:
Self-esteem maintenance in children. Journal of Personality and Social Psy-
chology, 46, 561-574.
Tuma, N. B., & Hallinan, M. T. (1979). The effects of sex, race, and achievement
in school children's friendships. Social Forces, 57, 1265-1285.
Vandell, D. L., & Mueller, E. C. (1980). Peer play and friendships during the first
two years. In H. C. Foot, A. J. Chapman, & J. R. Smith (Eds.), Friendship and

social relations in children (pp. 181-208). New York: Wiley.
Waas, G. A. (1987). Aggressive rejected children: Implications for school psy-
chologists. Jo urnal of School Psychology, 25, 383-388.
Weiss, R. S. (1973). Loneliness: The experience of emotional and social isolation.
Cambridge, MA: MIT Press.
Weiss, R. S. (1974). The provisions of social relationships. In A. Rubin (Ed.), Do-
ing unto others (pp. 17-26). Englewood Cliffs, NJ: Prentice-Hall.
Wentzel, K. R. (1991). Relations between social competence and academic
achievement in early adolescence. Child Development, 62, 1066-1078.
Wentzel, K. R. (1993). Motivation and achievement in early adolescence: The role
of multiple classroom goals. Journal of Early Adolescence, 13, 4-20.
Wigfield, A., & Eccles, J. S. (1992). The development of achievement task values:
A theoretical analysis. Developmental Review, 12, 265-310.
Williams, J. G., & Solano, C. H. (1983). The social reality of feeling lonely:
Friendship and reciprocation. Personality and Social Psychology Bulletin, 9,
237-242.
Young, J. G., Brasic, J. R., & Kisnadwala, H. (1990). Strategies for research on
school refusal and related nonattendance at school. In C. Chiland & J. G. Young
(Eds.), Why children reject school: Views from seven countries. The child and
his family, vol. 10 & Yearbook of the International Association for Child and
Adolescent Psychiatry for Allied Professions, (vol. 10, 199-223). New Haven:
Yale University Press.
5 Motivational approaches to aggression within
the context of peer relationships
Cynthia A. Erdley
For children in school, aggressive behavior often has adverse social and
academic consequences. Aggressive behavior negatively impacts upon
classroom functioning, as such behavior can be quite disruptive to the
learning environment, both for the individual and for the class as a whole.
Children who are aggressive are significantly more likely than their peers
to engage in off-task, disruptive classroom behavior, and this behavior has
been linked to these children experiencing academic difficulties (Coie &
Krehbiel, 1984). In addition, aggressive behavior is highly visible and
contributes greatly to these children earning an unfavorable image with
their teachers and peers (Coie & Koeppl, 1990). Indeed, children who are
excessively aggressive, hostile, and disruptive tend to be actively disliked,
or rejected, by their peers (Coie, Dodge, & Kupersmidt, 1990).
It is important to note that peer rejection is associated with a variety of
negative consequences for school children, including higher levels of
loneliness and social dissatisfaction (Asher, Parkhurst, Hymel, &
Williams, 1990) and problems with school transitions both into kinder-
garten (Ladd & Price, 1987) and junior high (Berndt, 1987). Furthermore,
children who are rejected by their peers are more likely to drop out of
school, to engage in criminal activity, and to suffer from mental health
problems (Kupersmidt, Coie, & Dodge, 1990; Parker & Asher, 1987). In-
terestingly, however, peer acceptance and aggression show somewhat dif-
ferent relations to school drop-out and criminality. Specifically, whereas
low peer acceptance is more predictive of dropping out of school than of
criminality, aggressiveness is more predictive of later criminality than of
dropping out of school (Parker & Asher, 1987). It is clear that aggressive
children have a negative impact on others, and that they themselves are at
risk for a variety of negative consequences, both in the short-term and the
long-term. An important question, then, is what motivates children to act
aggressively? An improved understanding of why children choose to be-
98
Motivational approaches to aggression 99
have aggressively seems to be crucial for the development of interventions

that would effectively decrease aggressive behavior.
Much of the research on children's peer relationships has examined the
link between children's behavior and their acceptance by peers (see Coie
et al., 1990, for a review). Increasingly, however, researchers in the area of
children's social behavior have been investigating how children who differ
in their level peer acceptance might vary in the kinds of cognitive process-
es they use when responding to social situations (see Dodge & Feldman,
1990, for a review). Although links between thoughts and acceptance have
been found, these links are typically not very strong. A potential explana-
tion for these relatively weak, though significant, relations is that social-
cognitive processing motivates certain behavior, and then children's be-
havior contributes to their acceptance level (Wentzel & Erdley, 1993).
Thus, it could be hypothesized that children's thoughts about the social
world would be more strongly related to children's behavior than to peer
acceptance, given the more proximal links between social-cognitive pro-
cessing and social behavior. Research has found that children who act ag-
gressively among peers approach social situations with certain thought
patterns that appear to motivate their aggressive behavior. This aggressive
behavior, as previously noted, then seems to lead to a variety of negative
consequences, such as low acceptance by peers.
In the following section, discussion focuses on how children's process-
ing of social information relates to aspects of their social adjustment, in-
cluding aggressive behavior and peer acceptance. A recent theoretical
model of social-information processing is presented, followed by a review
of empirical literature that provides support for this model.
Social information processing: Motivational links to

children's aggressive behavior
To enhance understanding of the kinds of thought processes that underlie
children's behavioral choices and the ways in which these social-cognitive
constructs might operate, several models of social information-processing
have been suggested (e.g., Dodge, 1986; Rubin & Krasnor, 1986). Crick
and Dodge (1994) have recently proposed a model of social information-
processing that is a reformulation of the model initially constructed by
Dodge (1986). According to this revised model, children approach a cer-
tain social situation with a database of memories of their past social expe-
riences, social schemas, and social knowledge. They then receive as input
a set of social cues, and their behavioral response is a function of how they
process those cues. These steps of processing include (1) encoding of ex-
ternal and internal cues, (2) interpretation of those cues, (3) selection of
goals, (4) response access, and (5) response decision.
Crick and Dodge's (1994) model of social information-processing sug-
gests several specific social-cognitive variables that may play a significant
role in motivating children's selection of aggressive behavioral responses.
One such variable is children's attributions of intent. As children encode
and interpret social cues during the first two steps of processing, they are
guided by relevant social knowledge that has been acquired through previ-
ous experiences. For example, children who have been frequently victim-
ized by peers are apt to attribute an act, such as being tripped by a peer, to
the peer's hostile intentions rather than to accidental circumstances. It
seems likely that children who believe that the other has harmed them on
purpose would be more motivated to respond to the provocateur in an ag-
gressive, retaliatory manner. Another important social-cognitive variable
is children's social goals, since it is hypothesized that during the third step
of social information-processing, children formulate a goal for the partic-
ular situation. Presumably, children who give priority to self-defense and
retaliation goals are likely to be motivated to engage in aggressive behav-
ior in their social interactions. The fourth step of processing involves the
variable of strategy knowledge, as children access possible behavioral re-
sponses to the situation from long-term memory. Children may be espe-
cially primed to implement aggressive responses if their social strategy
repertoire is mostly comprised of such responses.
Several social-cognitive constructs are likely to come into play in the
fifth step of social information-processing, when children must decide
upon a particular behavioral response. To select a certain response for en-
actment, children must feel confident that they can successfully produce
that behavior (i.e., feelings of self-efficacy). Furthermore, they must ex-
pect that positive outcomes will result from the behavior (e.g., If I push
that child off the swing, then I will get to use the swing myself). In addi-
tion, they must make a positive evaluation of the response based on their
moral rules or values (e.g., It's o.k. for me to push someone so that I can
get what I want). Therefore, children are likely to be especially motivated
to select an aggressive behavioral response if they think they are good at
enacting aggression, if they expect that positive outcomes will result if
they act aggressively, and if they believe that aggressive behavior is justi-
fied.
Using the Crick and Dodge (1994) model as a framework, I will focus
on four social-cognitive variables, attributions of intent, social goals, self-
efficacy perceptions, and beliefs about the legitimacy of aggression that

play significant roles in the sequence of social information-processing. I
will discuss how, together, these variables are likely to help in explaining
children's motivation to behave aggressively. It is important to note that
although Crick and Dodge propose that the path from a particular stimu-
lus (e.g., a single provocation by a peer) to a behavioral response (e.g., re-
taliation) logically follows a sequence of steps, they view social informa-
tion-processing as cyclical in nature, with each step of processing
influencing the others through a series of feedback loops. For example,
whereas children who interpret a peer's harmful behavior as intentionally
committed may be motivated to pursue a goal that focuses on retaliation, it
may also be that, through a feedback process, children who are primarily
concerned with self-defense goals may be more likely to interpret peers'
actions as hostilely motivated. Consequently, children's attributions,
goals, self-efficacy perceptions, and legitimacy of aggression beliefs are
not assumed to be independent of each other, and these four variables,
when added together, are believed to be predictive of aggressive behavior.
Although additional constructs (e.g., strategy knowledge, outcome expec-
tations) are also likely to have an impact on children's behavioral choices,
I will limit the discussion to these four social-cognitive variables.
It should be noted that this chapter is primarily concerned with intra-
personal processes that relate to social functioning. That is, these social-
cognitive variables are viewed as key motivational constructs that influ-
ence children's decisions regarding the enactment of aggressive behavior.
Nevertheless, it is important to recognize that the kinds of thought
processes in which children engage are likely to be impacted by various
social influences, including peers and the family. The ways in which peers
interact with children are apt to affect how children process information in
future social situations. For example, children who are repeatedly victim-
ized by peers are likely to develop a propensity to attribute behaviors that
harm them to their peers' hostile intentions. In addition, based on their
previous negative experiences with peers, these children may feel that it is
important to pursue social interaction goals that focus on retaliation. Par-
ents and/or siblings may also influence children's social information-pro-
cessing by encouraging children to take a certain kind of approach to their
social interactions. For example, children may be taught to suspect that
others usually have bad intentions and to believe that their own aggressive
behavior is justified. Thus, although the social-cognitive constructs repre-
sent intrapersonal processes, these processes are very likely to be subject
to interpersonal influences.
In the remainder of this chapter, I will review past research that has
studied how children who vary in their behavioral style and/or their level
of peer acceptance differ in their attributions about the intentions of oth-
ers, their social goals, their self-efficacy perceptions, and their beliefs
about the legitimacy of using aggression. Then, I will present some of my
recent work that examines the relative contributions of these four process-
es in explaining the aggressive behavior of boys and girls. Also investigat-
ed in this study is the question of how these social-cognitive processes,
along with children's aggressive behavior, relate to children's level of ac-
ceptance by peers. Finally, I will discuss implications for interventions
that try to decrease aggressive behavior in children. Specifically, I will de-
scribe the importance of targeting social-cognitive processes that might
reduce children's motivation to behave aggressively, as well as the poten-
tial difficulties in using such an approach.
Attributions of intent
Social situations are often unstructured, making it necessary for children
to interpret events and decide on their behavioral responses as they are in-
teracting with their peers (Renshaw & Asher, 1983). Different children
may respond in various ways to a certain social situation because of indi-
vidual differences in the attributions they make about that situation. Much
of the research on children's attributions of intent has been conducted by
Dodge and colleagues (e.g., Dodge, 1980; Dodge & Frame, 1982; Dodge
& Newman, 1981). In this work, children are presented with various so-
cial situations, including ambiguous provocation vignettes in which the
protagonist performs with ambiguous intentions some potentially provok-
ing act that has negative consequences for them. For example, children are
told to imagine that they are sitting in the lunchroom at school, and when
they are not looking, a peer spills milk on them. Results of these studies,
which involved boys only, have consistently shown that in response to am-
biguous provocation, when compared to nonaggressive, better-accepted
boys, aggressive, low-accepted boys are more likely to attribute hostile in-
tentions to the protagonist. This same tendency to interpret a protagonist's
actions as hostile has also been observed among aggressive, emotionally
disturbed boys (Nasby, Hay den, & DePaulo, 1980) and highly aggressive
adolescents (Graham, Hudley, & Williams, 1992; Slaby & Guerra, 1988).
In addition to attributing hostile intentions to the protagonist, aggres-
sive, low-accepted boys are more likely to report that they would respond
to the protagonist in an aggressive manner. Moreover, these boys expect to
be the recipients of more aggression in the future and mistrust the protag-
onist more (Dodge, 1980). Based on these findings, Dodge (1980) pro-
posed a cyclical relationship between attributions and aggressive behav-
ior. The cycle begins with an ambiguously-motivated action that
aggressive, low-accepted children attribute to the peer's hostile intentions.
This attribution "confirms" the children's general image of peers as being
hostile toward them and may increase their likelihood of interpreting fu-
ture harmful behaviors as negatively intended. After making a hostile at-
tribution, children will aggress against the peer, who may then aggress
back, further confirming for the aggressive children that others are pur-
posely harming them. This formulation meshes well with Berkowitz's
(1977) notion that aggressive behavior is governed by the interaction of
external conditions, such as behaviors observed being performed by oth-
ers, and internal qualities, including the perceiver's long-lasting habits and
personality traits or short-lived emotional states. Berkowitz argued that
environmental cues alone usually are not enough to cause aggressive be-
havior. Rather, the individual must make a cognitive appraisal of the
event, and the meaning given to the event will determine its cue properties
and thereby control its ability to elicit particular reactions from the per-
ceiver. Berkowitz claimed that some people are much more likely to de-
fine events as "aggression," so these events are more apt to precipitate re-
taliatory reactions in these individuals. The aggressive, low-accepted boys
studied in Dodge's research are a good example of such a group of people.
Interestingly, when investigating factors that might influence aggres-
sive, low-accepted boys' tendency to attribute hostile intent, Dodge and
Frame (1982) found that whereas these children tended to believe that the
protagonist had purposely caused harm when the negative actions or out-
comes were directed at them, they did not attribute hostility when another
peer was the victim of provocation. This suggests that these children have
a paranoid view of others rather than a cynical view, since they seem to
think that peers act with hostile intent specifically toward them, but not
generally toward other people.
Although aggressive, low-accepted children exhibit systematic differ-
ences in attributing hostile intent depending on the victim (self versus
peer), they continue to attribute hostility more than their nonaggressive,
better-accepted peers regardless of their feelings about the protagonist.
Fitzgerald and Asher (1987) presented children with a set of hypothetical
ambiguous provocation situations. In some situations the protagonist was
a peer highly liked by the subject, and in other situations the protagonist
was a peer highly disliked by the subject. Despite the fact that children
(both boys and girls participated in the study) attributed less hostility
overall to highly liked peers, the greater tendency to attribute hostile intent
still appeared in aggressive, low-accepted children, regardless of how well
liked the protagonist was. Thus, aggressive, low-accepted children seem
to believe that others act toward them in hostile ways, even when the pro-
tagonist is someone they report liking. Furthermore, they think the protag-
onist should receive harsh treatment.
Together, these findings demonstrate that children's aggressive behav-
ior appears to be motivated, at least in part, by their beliefs that others
bring harm to them intentionally. These beliefs may certainly have some
basis in reality, since aggressive children not only are more likely to initi-
ate aggression but also are more apt to be victims of aggression them-
selves (Dodge & Frame, 1982). Once children get into a cycle of expect-
ing hostility from their peers, responding aggressively, and then receiving
further negative treatment, this pattern is likely to contribute to the main-
tenance of their low acceptance within the peer group. Furthermore, this
tendency to interpret peers' actions as hostile is apt to detract from the
overall quality of any close friendships these children may have, since ag-
gressive, low-accepted children seem to be distrustful even toward highly
liked peers (Fitzgerald & Asher, 1987).
In summary, as some children approach their social world with beliefs
that others intend to harm them, this attributional style appears to increase
the likelihood that these children will be motivated to respond aggressive-
ly toward peers who provoke them. Notably, recent research by Graham,
Hudley, and Williams (1992) indicates that emotion, specifically anger,
plays a key role in the relation that has consistently been found between
hostile attributions and aggressive behavior. Their results suggest a se-
quence in which perceived intentionality elicits anger, and then anger mo-
tivates aggression. Thus, it seems that for children who experience anger
after attributing hostile intent to a peer, their anger is an important mecha-
nism in the triggering of an aggressive response.
Social goals
While children are interpreting social cues in a particular situation, they
must also select certain goals for that situation. Social goals are defined as
the intended outcomes that individuals are trying to achieve when inter-
acting with others (Pervin, 1989). Differences in the kinds of goals chil-
dren choose to pursue in their social interactions may help explain why
children confronted with the same type of social situation may respond in
quite different ways. For example, when playing a game, some children
may be mostly oriented toward winning and dominating others, while oth-
er children may be more concerned with having fun and developing per-
sonal relationships (Taylor & Asher, 1984). The former group would be
more likely to be aggressive and competitive, whereas the latter group
would be more apt to be friendly and supportive of others. Most of the re-
search on individual differences in social goals has focused primarily on
potential goal differences among children who vary in their level of peer
acceptance (e.g., Renshaw & Asher, 1983; Wentzel, 1991). It has been
proposed (Asher & Renshaw, 1981; Renshaw & Asher, 1982) that part of
the reason that low-accepted children engage in antisocial behavior is that
they are more likely to endorse antisocial goals in response to various so-
cial situations, such as those encountered in the school setting.
Several studies have found that low- versus better-accepted children do
differ in the kinds of social goals they reportedly pursue. Renshaw and
Asher (1983) presented third- through sixth-grade children with hypothet-
ical situations involving initiating interactions with peers, dealing with
peer rejection, maintaining friendships, and handling interpersonal con-
flict. Low-accepted children were no more likely than their better-accept-
ed peers to suggest avoidance goals or hostile goals for these situations.
However, the low-accepted children proposed significantly fewer proso-
cial goals. Thus, the low-accepted children did not openly endorse antiso-
cial goals. However, they were less likely to generate prosocial, relation-
ship-oriented goals for these situations, suggesting that they are less
positive and friendly in their social orientation than are better-accepted
children.
In another study, Wentzel (1991) assessed the goal patterns of sixth-
and seventh-grade children of various levels of acceptance by asking them
how often they tried to pursue different outcomes. These included a set of
prosocial goals (e.g., "How often do you try to help other kids when they
have a problem?") and a set of social interaction goals (e.g., "How often
do you try to be with other kids rather than by yourself?"). Low-accepted
children reported trying to achieve the prosocial and social interaction
goals significantly less often than did the better-accepted children. These
results suggest that low-accepted children place lower priority on relation-
ship-oriented goals, since they appear to pursue such goals less frequently
than do their better-accepted peers.
In a study that has implications for children's social goals, Crick and
Ladd (1990) examined third- and fifth-grade children's outcome expecta-
tions, which were defined as how children expected peers and adults to re-
act to them in social situations. Crick and Ladd found that when compared
with their better-accepted peers, low-accepted children more often fo-
cused on whether a particular strategy would result in the accomplishment
of some instrumental goal (e.g., attaining compliance to a command)
rather than on how the strategy might influence the relationsip between
the self and the peer. Crick and Ladd proposed that low-accepted chil-
dren's greater attention to instrumental outcomes suggests that these chil-
dren prioritize their social goals in such a way that they are more interest-
ed in pursuing an instrumental goal and will give this goal priority even at
the expense of maintaining a relationship. Recent work by Rabiner and
Gordon (1992) has shown that, indeed, low-accepted children tend to fo-
cus on the exclusive attainment of instrumental goals, even after they have
been explicitly encouraged to integrate instrumental and relational goals.
Together, these studies indicate that low-accepted children are less pos-
itive and friendly in the social goals they pursue. Although modest goal
differences have been found among children of different levels of peer ac-
ceptance, it seems that in order to determine whether various goal orienta-
tions underlie differences in behavior, more emphasis should be placed on
the relation between goals and behavior. Although it is true that low-ac-
cepted children are more antisocial in their behavioral style, not all low-
accepted children are aggressive. Indeed, some low-accepted children
exhibit socially withdrawn behavioral styles (Coie et al., 1990). Conse-
quently, not all low-accepted children would be expected to endorse hos-
tile goals. In fact, perhaps by combining low-accepted children into one
group, regardless of their behavioral style, potential goal differences in
behavioral subgroups of low-accepted children have been obscured. It
may be that aggressive children are more likely to pursue hostile goals,
whereas behaviorally withdrawn children are more apt to endorse socially
avoidant goals but not hostile goals. As previously suggested, it seems
likely that children's social goals are likely to motivate their behavior, and
then their behavioral style will contribute to their level of acceptance by
peers. To understand the kinds of goals that might specifically motivate
aggressive behavior, it seems necessary to examine the social goals en-
dorsed by aggressive children. In one such study, Slaby and Guerra (1988)
found that, in a forced-choice task, high-aggressive adolescents were
more likely than low-aggressive adolescents to select a hostile social goal
than a nonhostile goal in response to ambiguous provocations.
A broader understanding of the hostile motivations that are likely to in-
fluence aggressive children's selection of antisocial goals is provided by
research on the values aggressive children place on the rewarding out-
comes of aggression (Boldizar, Perry, & Perry, 1989). This work suggests
that aggressive children might especially value the goals of self-defense,
retaliation, and control over the protagonist. At the same time, aggressive
children ascribe relatively less value to peer interactions and peer feelings,
suggesting that they would be less likely to endorse relationship-oriented
goals. In a recent study, Lochman, Wayland, and White (1993) asked ado-
lescent boys to rate the importance of pursuing four different kinds of
goals in response to a single ambiguous provocation. Compared to nonag-
gressive boys, aggressive boys rated the goals of revenge and dominance
high in value and the goal of affiliation low in value. There were no differ-
ences between aggressive and nonaggressive boys in the value they as-
cribed to the socially avoidant goal. Furthermore, those who rated the
dominance and revenge goals high and the affiliation goal low were more
likely to report crime against persons (e.g., assault) as well as higher lev-
els of marijuana, drug, and alcohol involvement. They also suffered from
lower self-esteem and higher levels of peer rejection. Thus, those who give
high priority to hostile social goals are also more likely to have a variety
of negative social experiences.
Erdley and Asher (in press) have also examined the kinds of goals that
children might pursue following ambiguous provocation. The fourth- and
fifth-grade participants (both boys and girls) rated the extent to which
they agreed or disagreed that they would be trying to accomplish various
goals following provocation. The collection of goals included some retali-
ation goals, avoidant goals, and relationship-oriented, prosocial goals.
Children were categorized as aggressive, withdrawn, or prosocial based
on their primary behavioral response across ten ambiguous provocation
situations, and they made their goal ratings in reaction to three of those
situations for which they had given their predominant behavioral re-
sponse. It was found that compared to the prosocial and withdrawn re-
sponders, the aggressive responders gave higher ratings to the retaliation
goals of trying to get back at the other child and trying to make the other
child feel bad. The aggressive responders also rated higher the goals of
looking strong and protecting themselves. The aggressive responders were
less concerned with the socially avoidant goal of trying to stay away from
the protagonist. Finally, compared to the prosocial and withdrawn respon-
ders, the aggressive responders gave lower ratings to the prosocial goals of
trying to work things out peacefully, trying to get along with the other
child, and trying to take care of the problem created by the protagonist.
Therefore, it appears that when provoked, those who endorse aggressive
responses put more emphasis on punishing the protagonist and defending
the self, and they are less concerned with trying to preserve a positive re-
lationship with the protagonist or deal with the problem in a constructive
manner. Together, the studies that have examined the relation of children's
social goals to their behavioral style suggest that children who give priori-
ty to hostile social goals, especially at the expense of prosocial goals, are
more likely to be motivated to engage in aggressive interactions with
peers.
Self-efficacy perceptions
In addition to the social goals children pursue, children's confidence re-
garding their effectiveness in being able to carry out certain actions is apt
to influence their chosen behavioral responses to various social situations.
Bandura (1981) defined self-efficacy as individuals' beliefs about their
abilities to execute and regulate the important actions in their lives, and
such self-perceived competencies affect the person's choice of what activ-
ities to undertake versus avoid. It seems quite reasonable to predict that
children's goals and self-efficacy perceptions would be closely related, as
individuals are likely to pursue those goals they feel most confident in at-
taining (see Schunk & Zimmerman, this volume).
Crick and Ladd (1990) proposed that children who are rejected by their
peers focus more on instrumental goals because, based on previous social
experiences, these children may feel helpless about their ability to fulfill
relationship-oriented goals successfully. Such children primarily pursue
instrumental goals because they are more confident about achieving suc-
cessful outcomes. In contrast, children who are better-accepted by their
peers predominantly pursue relationship-oriented goals. This is the case
because, as a result of their generally positive interactions with peers, bet-
ter-accepted children are likely to feel fairly confident about their ability
to fulfill prosocial goals.
To examine how children's self-efficacy perceptions are related specif-
ically to their behavior, as opposed to their level of peer acceptance, sever-
al studies have been conducted investigating the perceptions of children
who differ in their behavioral styles. Perry, Perry, and Rasmussen (1986)
compared children who were described by peers as either aggressive or
nonaggressive on their ratings of how easy or difficult it would be for
them to accomplish various prosocial and antisocial actions. Perry et al.
found that aggressive children reported that it was easier to perform ag-
gression and more difficult to inhibit aggressive reactions. Aggressive
children were also more confident that they could achieve tangible re-
wards and reduce aversive treatment by others via aggression. There was
no evidence that aggressive children experienced more difficulty in per-
forming prosocial behaviors than did nonaggressive children. Neverthe-
less, these results suggest that aggressive children may be quite motivated
to enact aggressive responses because they are fairly confident about us-
ing such behavior and expect their aggression to lead to positive outcomes
for themselves.
In a recent study, Erdley and Asher (in press) investigated the self-effi-
cacy perceptions of fourth- and fifth-grade children who varied in their
predominant behavioral response (aggressive, withdrawn, or prosocial) to
ten ambiguous provocation situations. In reaction to a subset of three of
the situations (to which they had given their primary behavioral response),
children rated the extent that they agreed or disagreed that they would be
good at accomplishing various goals if they tried. Compared to the proso-
cial and withdrawn responders, the aggressive responders were more con-
fident that they would be good at getting back at the other child, making
the other child feel bad, and looking strong. Interestingly, the aggressive
responders rated their effectiveness in staying away from the protagonist
lower than did the other behavioral response groups. Finally, compared to
the prosocial and withdrawn responders, the aggressive responders were
less confident that they would be good at working things out peacefully
with the protagonist, getting along with the other child, and taking care of
the problem created by the protagonist.
Like the Perry et al. (1986) study, the Erdley and Asher (in press) study
found that aggressive children were relatively more confident in their anti-
social abilities. However, unlike Perry et al.'s results, in which aggressive
and nonaggressive children did not differ in their confidence regarding
their prosocial abilities, the Erdley and Asher study found that aggressive
children felt less efficacious about their prosocial skills than did their
nonaggressive counterparts. These discrepant findings may be explained
by considering the situations used when assessing children's self-efficacy
perceptions. In the Perry et al. study, children were asked to evaluate their
prosocial abilities in situations that did not involve direct threat to the self
(e.g., children rated how easy or difficult it would be for them to help a
child who has a broken arm). The ambiguous provocation situations em-
ployed by Erdley and Asher may have aroused defensive reactions in the
aggressive responders and consequently may have decreased aggressive
responders' confidence in being able to use prosocial skills under these
more emotionally-charged circumstances. The results from the Erdley and
Asher study suggest that certain children are motivated to respond aggres-
sively to provocation, not only because they think they could successfully
enact aggressive behaviors, but also because they believe they are less
skilled at performing prosocial actions. Interestingly, aggressive children
reported that it would be relatively difficult for them to stay away from the
protagonist (Erdley & Asher, in press) and that they would have problems
inhibiting aggression (Perry et al., 1986). Consequently, it seems that es-
pecially when provoked, aggressive children have difficulty avoiding a
confrontation, and as they approach the protagonist, it is more likely to be
with hostile rather than prosocial motivations. Thus, it appears that chil-
dren's aggressive behavior is influenced by their perceptions that they can
use aggression successfully and that they are less skilled at enacting
nonaggressive responses.
Legitimacy of aggression beliefs

An additional factor that is likely to impact upon children's decision about
whether to act aggressively is their beliefs about the acceptability and le-
gitimacy of aggression as a response to social situations. It seems quite
plausible that people who vary in their beliefs about the appropriateness
or moral correctness of aggression would differ in their levels of aggres-
sive behavior. Indirect evidence for this point comes from several studies.
Bandura (1979, 1986) reported that individuals who aggress typically jus-
tify their negative actions in various ways, such as blaming the victim. In
other work investigating reactions to aggressive behavior, Boldizar et al.
(1989) asked children to respond to hypothetical situations in which they
directed aggressive acts toward peers. Children identified by peers as ag-
gressive were less apt than nonaggressive children to make negative self-
evaluations following aggressive behavior. In another study, Perry and
Bussey (1977) gave children the opportunity to reward themselves after
being led to believe that they had hurt another child. It was found that ag-
gressive children rewarded themselves more than did nonaggressive chil-
dren. Thus, for aggressive children, the injury of another child elicited
positive self-evaluations. These studies point to the strong possibility that
aggressive children believe that their aggressive behavior is justified.
Even more direct evidence concerning children's beliefs about the ap-
propriateness of aggression comes from a study of 15-to 18-year-old ado-
lescents by Slaby and Guerra (1988). Slaby and Guerra administered a
true/false measure that assessed individuals' beliefs about whether aggres-
sion is acceptable (e.g., "It's o.k. to hit someone if he or she hits you
first"). Adolescents identified by teachers as highly aggressive were more
likely than their nonaggressive counterparts to believe that aggression is

legitimate. In a more recent study, Erdley and Asher (1993) measured the
legitimacy of aggression beliefs of fourth- and fifth-grade students. The
questionnaire was a modification of the Slaby and Guerra (1988) measure
(e.g., simplified wording of items, use of a rating scale rather than
true/false responding). Children's behavior was assessed using both self-
reports of behavior, specifically children's responses to hypothetical situa-
tions involving ambiguous provocation, and peer evaluations of children's
aggressive, withdrawn, and prosocial behavior. Results from both mea-
sures of behavioral style showed that children who strongly believe in the
legitimacy of aggression are more likely to engage in aggression and less
apt to be prosocial or withdrawn with peers. Thus, it is clear that children's
beliefs about aggression are related to their behavior. Those children who
view aggressive behavior as justified seem to be particularly motivated to
act aggressively and are less likely to engage in alternative, nonaggressive
reponses.
A study of social-cognitive processes, aggressive behavior,

and peer acceptance
From the above review, it appears that several social-cognitive variables
are associated with aggressive behavior. Aggressive responses are at least
partially motivated by children's attributions that others have intentionally
harmed them, by their endorsement of hostile goals, by their perceptions
that they can effectively enact aggressive responses, and by their beliefs
that aggression is legitimate. The model proposed earlier in this chapter is
that children's thoughts about the social world motivate their behavior, and
their behavior then contributes to their level of acceptance by peers. The
research that has been done to date has mostly focused either on how chil-
dren's behavior relates to their acceptance level, how their social-cognitive
processes relate to their acceptance level, or how their social-cognitive
processes relate to their behavior. Rarely have all three factors been inves-
tigated in one study (see Wentzel & Erdley, 1993, for an exception). In ad-
dition, many of the studies (e.g., Dodge, 1980; Dodge & Frame, 1982;
Dodge & Newman, 1981; Lochman, Wayland, & White, 1993) have in-
cluded boys only.
To address some of the limitations of previous research, I conducted a
study in which measures of boys' and girls' social-cognitive processes,
self- and peer-assessed aggressive behavior, and level of peer acceptance
were obtained. In response to six ambiguous provocation situations, 506
fourth-, fifth-, and sixth-grade students made attributions about the intent
of the peer provocateur and rated their likelihood of reacting aggressively
to the peer. Also in response to these situations, children rated the extent
to which they would be trying to accomplish each of several aggressive
goals (i.e., trying to get back at the other child, trying to make the other
child feel bad, trying to look strong). Children then evaluated their self-ef-
ficacy for aggressive behavior by rating the degree to which they thought
they would be good at accomplishing each of these goals if they were to
try to achieve the goal. Next, children completed a questionnaire (Erdley
& Asher, 1993) in which they indicated their level of agreement with
items concerning the legitimacy of aggression (e.g., "It's o.k. to hit some-
one if that child really makes you angry"; "It's o.k. to say something mean
to someone if that child does something mean to you"). In addition, peer
evaluations of children's aggressive behavior in everyday school life were
obtained, using Rockhill and Asher's (1992) aggression subscale (i.e.,
"Who starts fights?" "Who is mean?," "Who gets mad easily?," and
"Who hits, pushes, or kicks?"). Finally, to measure children's level of ac-
ceptance by the peer group, students were asked to rate how much they
like to play with each of their classmates.
In the following sections, results concerning the relations among chil-
dren's social-cognitive processes, aggressive behavior, and peer accep-
tance level are presented. All analyses were conducted separately for boys
and girls to investigate whether the relations of the social-cognitive
processes to aggressive behavior and acceptance vary by gender.
Social-cognitive processes and aggressive behavior

for boys and girls
Table 5.1 shows correlations among attributions of intent, aggressive
goals, efficacy for aggression, legitimacy of aggression beliefs, aggres-
sive behavior (both self- and peer-reported), and peer acceptance for boys
and girls. For both sexes, all of the social-cognitive processes regarding
aggression are significantly correlated with one another. Thus, it seems
that children create multiple patterns of beliefs that are associated with
each other and that promote (or argue against) their enactment of aggres-
sive behavior. For both boys and girls, aggressive social goals show the
strongest correlations with the other social-cognitive processes. For exam-
ple, it appears that children who are especially motivated to retaliate in so-
cial interactions are also more likely to define situations in ways that cre-
ate for them the need to aggress. In other words, they are more apt to think
Table 5.1. Intercorrelations among social-cognitive processes, aggressive be-

havior, and peer acceptance for boys and girls
Variable 1 2 3 4 5 6 7
1. Attributions - 49** .23** .40** .54** .16* -.06
2. Goals .50** - .71** .63** .84** .20** -.03
3. Efficacy .28** .66** - .46** .54** .10 .02
4. Legitimacy .31** .61** .38** - .65** .09 .01
5. Aggression .56** .81** .51** .64** - .21** -.07
(Self-reported)
6. Aggression .08 .02 .02 .04 .11 - -.46**
(Peer-assessed)
7. Acceptance .01 -.10 .01 -.10 -.10 -.32** -
Values above the diagonal are for boys (n = 247), and values below the diagonal are for girls (n =
259). **/?<.O!,*/?<.05.
that others harm them intentionally and to pursue aggressive goals. It is

important to note, however, that these data cannot be interpreted in terms
of causality. It may be that children's attributions of hostile intent lead
them to pursue aggressive goals, or that children's aggressive goals con-
tribute to their hostile attributional style. Nevertheless, children's aggres-
sive goals and attributional processes are significantly related to one an-
other. In addition, the endorsement of aggressive goals is strongly
correlated with children's perceptions that they are good at achieving
these goals. This finding supports the proposition that children are espe-
cially likely to pursue those goals that they think they would be able to ac-
complish. Children's perceptions that they can effectively fulfill aggres-
sive goals are likely to encourage their pursuit of these goals, particularly
if they think that there is justification for endorsing these goals. In fact,
children's propensity to pursue retaliation goals is significantly associated
with beliefs that such hostile exchanges are legitimate. Therefore, children
who tend to endorse aggressive goals appear to maintain multiple thought
patterns that might contribute to their enactment of aggressive behavior.
These findings suggest that various social-cognitive processes work to-
gether as potential motivators of children's aggressive behavior. The next
question, then, is are these beliefs actually associated with children's ag-
gressive behavior? The correlational analyses in Table 5.1 show that for
both boys and girls, attributions of intent, aggressive goals, efficacy for
aggression, and legitimacy of aggression beliefs are significantly associ-
ated with their self-reported aggressive responses to ambiguous provoca-
tions. However, in terms of children's general aggressive behavioral char-
acteristics, as assessed by peers, for boys, only attributions of intent and
aggressive goals are associated with their behavior. Interestingly, for girls,
none of the social-cognitive processes are associated with their general
aggressive behavior.
To examine the relative contributions of these processes in explaining
boys' and girls' aggressive behavior, aggressive behavior was regressed on
the four social-cognitive variables. Separate analyses were conducted for
self-reported and peer-assessed aggression. As can be seen in Table 5.2,
for both boys and girls, attributions of intent, aggressive goals, and legiti-
macy of aggression beliefs were independent, significant predictors of
self-reported aggressive behavior. Notably, aggressive goals were the
strongest predictor of self-reported aggressive behavior for both boys and
girls. When predicting peer-assessed aggressive behavior, for boys, ag-
gressive goals were the only independent predictor, and for girls, none of
the social-cognitive processes made an independent, significant contribu-
tion.
Together, these findings suggest that for both boys and girls, social-
cognitive processes are strongly associated with their likelihood of report-
ing aggressive behavior in response to a specific kind of situation, espe-
cially a situation (ambiguous provocation) that is likely to elicit
aggression in some individuals. That is, children's aggressive responses to
these provocations appear to be closely related to how they view these par-
ticular situations, what they want to accomplish in these social interac-
tions, and whether they think aggressive behavior is justified. However,
children's social-cognitive processes regarding aggression show weak re-
lations to their overall aggressive behavior, as evaluated by peers. One ex-
planation for this finding is that peers assessed children's general level of
aggressive behavior, presumably making judgments across a wide variety
Table 5.2. Results of simultaneous regressions of social-cognitive processes

on the aggressive behavior of boys and girls
Attributions Aggressive Efficacy for Legitimacy of
of Intent Goals Aggression Aggression R2
Boys' aggression
Self-reported .15** .66** -.08 .17** .74**
Peer-assessed .08 .20** -.08 -.06 .05*
Girls' aggression
Self-reported .20** .58** -.01 .23** .73**
Peer-assessed .09 -.08 .03 .05 .01
Standardized beta weights are shown. Sample sizes for boys and girls are 247 and 259, respec-
tively. **/?<.01, *p<.05.
of situations. And, it would be expected that children's self-reported

thoughts in response to a specific kind of event that is directly related to
aggression would be more strongly associated with their behavior in that
particular situation than with more global aspects of their behavior.
A second explanation for the weaker findings regarding the associa-
tion between social-cognitive processes and peer-assessed aggressive be-
havior is that peers' evaluations are based on their observations and in-
terpretations of children's actual behaviors, whereas children are focused
on their planned course of action. Consequently, children may report that
they would respond aggressively, but they may not really do so in actual
social situations, or they may do so, but in a way that is not recognized
by peers as aggressive. Interestingly, the correlation between girls' self-
reported and peer-assessed aggressive behavior was not significant (r =
.11, n.s.), but the correlation for boys was (r = .21, p < .01). This sug-
gests that for girls, there is not a relation between what their behavioral
objectives are in response to ambiguous provocation and what their peers
observe regarding their actual aggressive behavior. This may help to ex-
plain why girls' thoughts regarding aggression were not found to be re-
lated to their overall aggressive behavior, although these social-cognitive
processes significantly predicted to their self-reported aggressive behav-
ior. Girls' thoughts about aggression do seem to motivate their aggres-
sive behavior, but not necessarily in a way that is recognized as aggres-
sive by peers. In contrast, for boys, there appears to be more consistency
in their aggressive behavior across contexts and in how their social-cog-
nitive processes relate to their aggressive behavior. Specifically, boys'
self-reported aggressive behavior in response to ambiguous provocations
is significantly related to their general aggressive behavior patterns, as
evaluated by peers. Furthermore, boys' thought processes regarding ag-
gression seem to motivate their aggressive behavior, both in terms of
what the boys themselves report they would do and in terms of the be-
havioral characteristics that peers observe.
Gender differences in motivational systems

regarding aggressive behavior
It is also possible that boys' and girls' social-cognitive processes regarding
aggression may motivate them to engage in different forms of aggressive
behavior, based on their higher-order social goals. In fact, the physical and
verbal forms of aggression that were assessed in the present study appear
to be more relevant to the social lives of boys than of girls. Whereas ag-
gressive behavior has been defined as behavior that is intended to hurt or

harm others (Berkowitz, 1993), Crick and Grotpeter (1995) have proposed
that when attempting to inflict harm on peers, children do so in ways that
best thwart or damage the goals that are valued by their respective gender
groups. The goals of instrumentality and dominance have been shown to
be important to boys, whereas relationship-oriented goals are more valued
by girls (see Block, 1983, for a review). Crick and Grotpeter (1995) found
that boys are more likely to harm others through physical and verbal ag-
gression, whereas girls are more apt to engage in relational aggression,
i.e., behaviors that are intended to damage another child's friendships, or
feelings of inclusion by the peer group. Although research has consistent-
ly shown that boys are more aggressive than girls (see Parke & Slaby,
1983, for a review), Crick and Grotpeter (1995) have found that this gen-
der gap in aggression narrows when relational as well as overt forms of
aggression are assessed.
It appears that gender differences in the enactment of aggressive be-
havior exist because different motivational systems may be operating for
boys and girls. Perhaps when boys endorse aggressive goals, they are also
influenced by their higher-order goals of dominance and instrumentality
as they decide how to fulfill their aggressive goals. If it is particularly im-
portant for boys to assert themselves and control peers across a variety of
social situations, then they are likely to try to accomplish these goals
through physical and verbal aggression. Therefore, it is likely that signifi-
cant relations were found between boys' social-cognitive processes and
their general aggressive behavior in the present study because the forms
of aggression that were evaluated by peers are in fact more characteristic
of boys' aggressive behavioral styles.
It is possible that when girls endorse aggressive thought patterns, they
are also influenced by their higher-order goals of affiliation with peers.
Consequently, in actual social interactions, girls may be more likely to act
on their aggressive motivations by using relational aggression. For exam-
ple, girls may choose to accomplish their aggressive goals by excluding a
peer from participating in a group rather than by hitting the peer. Thus,
perhaps part of the reason that girls' peer-evaluated aggressive behavior
was not significantly predicted by their social-cognitive processes is that
forms of aggression that are more relevant to girls' peer groups were not
assessed in the present study. It may be that girls' social-cognitive process-
es are more closely tied to their enactment of relational aggression, since
this form of aggression is more characteristic of them.
Relations of peer acceptance to social-cognitive processes and

aggressive behavior for boys and girls
As can be seen in Table 5.1, neither boys' nor girls' social-cognitive
processes were significantly correlated with their peer acceptance level.
However, for both boys and girls, aggressive behavior was significantly
and negatively correlated with acceptance. Furthermore, when acceptance
was regressed on attributions of intent, aggressive goals, efficacy for ag-
gression, legitimacy of aggression beliefs, and aggressive behavior, ag-
gressive behavior was the only significant, independent predictor of ac-
ceptance for both boys (B = -.46, p < .01) and girls (B = -.32, p < .01).
Thus, it appears that there are not direct associations between thought
processes and acceptance for either boys or girls. These results suggest a
pathway in which social-cognitive processes motivate aggressive behav-
ior, and then aggressive behavior contributes to reduced acceptance by
peers. It may be that even stronger relations between aggression and ac-
ceptance could be found, particularly for girls, if both overt and relational
forms of aggression are taken into account. It seems important that future
work on motivational patterns that contribute to aggression examine the
association of social-cognitive processes with relational aggression, espe-
cially to better understand girls' aggression. Furthermore, it would be in-
teresting to investigate the relative contributions of overt and relational
aggression in predicting boys' and girls' peer acceptance.
Implications for intervention

Both past research and the present study indicate that children's social-
cognitive processes regarding aggression are related to their aggressive
behavior. Furthermore, aggressive behavior is associated with negative
consequences for children in school, including academic difficulties and
social rejection by peers. Thus, it appears that to try to improve aspects of
children's school adjustment, efforts must be made to decrease their ag-
gressive behavior. One model that has been proposed by peer relationships
researchers is that children who exhibit antisocial behavior and who are
rejected by peers suffer from social skills deficits (Asher & Renshaw,
1981). Consequently, in many intervention studies (e.g., Ladd, 1981;
Oden & Asher, 1977), children have been taught various prosocial skills
(e.g., cooperation, communication, participation) thought to promote pos-
itive social interactions and peer acceptance. The social skills deficits
model seems to suggest that by teaching children prosocial skills, their an-
tisocial behavior will be replaced by more adaptive social strategies.
However, it appears important not only that prosocial skills be encouraged
but also that antisocial behavior be actively discouraged (Coie & Koeppl,
1990). Indeed, in intervention work by Bierman, Miller, and Stabb (1987),
children who received instruction in prosocial behavior as well as prohibi-
tions against negative behavior displayed greater increases in prosocial
behavior and decreases in antisocial behavior than children who received
only one aspect of the training.
Although past intervention efforts have been fairly successful, not all
children have shown improvements in behavior or gains in acceptance.
One possible explanation for why these programs have not been more ef-
fective is that these interventions have focused primarily on changing chil-
dren's behavior, and they have not addressed the social-cognitive variables
that seem to play a significant role in motivating maladaptive social be-
haviors. It seems reasonable to expect that after being repeatedly rein-
forced for not being aggressive and being taught to use prosocial skills,
children may begin to act less aggressively and more prosocially among
peers. However, if these children retain the same social-cognitive patterns
that initially motivated their aggression (e.g., a hostile attributional style),
they are likely to return soon to a pattern of aggressive interaction with
their peers.
To date, very few studies have attempted to decrease children's aggres-
sion by intervening at the social-cognitive level. However, the existing
work provides evidence that such interventions can be effective in chang-
ing children's thought processes regarding aggression and in reducing
their aggressive behavior. In a recent study, Hudley and Graham (1993)
conducted an intervention program that was designed to reduce aggressive
males' tendency to attribute hostile intentions to peers following ambigu-
ous provocation situations. Specifically, through discussions and role play
across twelve sessions, these boys were taught to identify ambiguous peer
intent, to make attributions to nonhostile intent when negative social en-
counters were portrayed as ambiguous, and to generate nonaggressive re-
sponses given attributional uncertainty. It was found that following the at-
tributional intervention, in response to both hypothetical situations and
laboratory simulations of ambiguous provocation, aggressive boys were
less likely to presume hostile intent by peers and less likely to react with
aggression. In addition, intervention subjects were rated as less aggressive
by their teachers following the treatment. Thus, it appears that boys' attri-
butional styles can be modified, and such alterations in thoughts do effec-
tively decrease their aggressive behavior, even in more generalized social

contexts.
An intervention study by Guerra and Slaby (1990) likewise supports
the proposition that modifying individuals' beliefs about aggression can
change their behavior. In this study, adolescents incarcerated for offenses
involving aggression participated in a twelve-session intervention that fo-
cused on several cognitive factors identified as correlates of aggression,
including beliefs about the legitimacy of aggression. The results of the in-
tervention were encouraging, as those adolescents who participated
showed decreased endorsement of beliefs supporting aggression, de-
creased aggressive behavior, and increased social problem solving skills.
However, following their release from the institution, the treatment groups
did not have a significantly lower recidivism rate than did the nontreat-
ment control groups. Therefore, although the intervention was effective in
the short-term, it appears that its influence may have diminished in the
long-term, perhaps because the adolescents returned to an environment
where their aggression was once again encouraged.
Whereas past intervention efforts have focused on children's attribu-
tions of intent and beliefs about the legitimacy of aggression, results of
the present study suggest that children's aggressive goals are likely to be
an important target for intervention when attempting to reduce children's
aggression. It appears that children who are especially focused on trying
to retaliate against peers or dominate the social situation are particularly
likely to engage in aggressive behavior. This suggests that attempts need
to be made to restructure the social goal hierarchy of aggressive children.
Such an intervention might take the form of addressing children's percep-
tions of the relative value of aggressive and prosocial goals and emphasiz-
ing the greater positive consequences of pursuing prosocial as opposed to
aggressive goals. Moreover, it is very important that such lessons be ac-
companied by practice in pursuing these prosocial goals so that children
can personally experience the benefits of giving priority to prosocial over
aggressive goals. In an intervention study with unpopular children that
may have indirectly influenced children's social goal priorities, Oden and
Asher (1977) coached children on several prosocial skills (e.g., coopera-
tion) before these children played games with peers. These skills were dis-
cussed during the coaching sessions and were also described as ways to
make the games more fun to play. Thus, it seems that the coaching not
only provided children with lessons about important prosocial skills, but
also encouraged children to pursue a prosocial goal (having fun) in the
game-playing situation rather than an aggressive goal, such as dominance.
It is possible that by focusing children on prosocial goals, and giving them

the opportunity to have positive social experiences when pursuing these
goals, children may eventually assign higher priority to prosocial goals in
their social interactions.
Furthermore, it appears that part of the reason that children pursue ag-
gressive goals is that they think they are good at achieving these goals,
and they believe that they are not as effective in fulfilling prosocial goals,
especially when they are provoked (Erdley & Asher, in press). It may be
that if children have the opportunity to engage in prosocial interactions in
a controlled social environment (e.g., a laboratory situation), their confi-
dence in their prosocial skills may increase, and they may then be more
likely to adopt prosocial goals in other contexts. It seems particularly im-
portant that children be given the opportunity to practice their prosocial
behaviors in response to situations that are potentially emotionally arous-
ing, such as ambiguous provocation or social failure, since these kinds of
situations seem to be quite problematic for aggressive children (Dodge,
McClaskey, & Feldman, 1985). Perhaps children's motivational patterns
for aggression are especially primed because they perceive these situa-
tions as very threatening. Hence, if children's aggressive motivations can
be successfully reduced under these types of circumstances, it is likely
that their overall level of aggressive behavior would decrease as well.
Clearly, it must be recognized that modifying social-cognitive process-
es regarding aggression is likely to be quite challenging. This is particu-
larly so because across development, children's experiences with various
socialization influences, including the family and peers, are likely to con-
tribute to the development and maintenance of certain social information-
processing patterns. For example, children who are in a family environ-
ment that endorses antisocial behavior and promotes thoughts that support
aggression are likely to receive repeated reinforcement for their more hos-
tile approach to social interactions (Pettit, Dodge, & Brown, 1988). More-
over, in the peer context, it has been found that aggressive children not
only are more likely to make hostile attributions and to react aggressively
to peers but also are more likely to be victims of aggression. Thus, their
perceptions of peers as hostile do have a basis in reality and are often con-
firmed through their experiences (Dodge & Frame, 1982). As a result of
their frequent and regular use of hostile attributions, it appears that ag-
gressive children's perception that others intend to harm them is a causal
construct that remains quite active in their memories. This construct is
then readily accessible in subsequent situations of attributional ambiguity,
and consequently aggressive children are primed to attribute hostile intent
(Graham & Hudley, 1994). In addition to their hostile perceptions of the

social world, aggressive children are likely to maintain positive beliefs re-
garding their own use of aggression because for many of these children,
such behavior results in a variety of instrumental rewards (Perry et al.,
1986), Therefore, it may be somewhat difficult to convince aggressive
children that prosocial behavior is more beneficial overall than is aggres-
sive behavior.
Together, these findings suggest that as children become older, the
ways in which they process social information are likely to become solidi-
fied through years of social experiences and may become increasingly re-
sistant to change. Perhaps future intervention efforts would be potentially
more successful, particularly in the long term, if they were attempted with
children at earlier ages (i.e., preschool) before thought patterns become
relatively stable (see Zahavi & Asher, 1978, for a relevant intervention ef-
fort). Family-based interventions during the preschool years may be espe-
cially effective in preventing children from developing social-cognitive
patterns that contribute to aggressive behavior. Moreover, peer-based in-
terventions should take into account the possibility that, regardless of their
age, aggressive children mostly associate with certain peers who encour-
age their aggression. Therefore, it may be important to conduct interven-
tions with groups of children who might then reinforce one another's
prosocial behavior and discourage aggressive behavior and the related ag-
gressive thought patterns.
Children's aggressive behavior appears to be motivated by several so-
cial-cognitive processes that are related to one another. Thus, it may be
that by modifying one aspect of children's motivation to act aggressively
(e.g., their aggressive goals), other aspects of their thoughts that might en-
courage aggressive behavior (e.g., their propensity to attribute hostile in-
tentions) may also change. However, it seems that the most effective so-
cial-cognitive interventions should try to target the variety of thought
processes that are associated with aggressive behavior. That is, by simulta-
neously trying to alter children's attributions of intent, their endorsement
of aggressive goals, their perceptions of efficacy for acting aggressively,
and their beliefs that aggression is legitimate, children's motivation to en-
gage in aggressive behavior may be decreased. Reducing the incidence of
aggressive behavior of children in school is especially important because
such behavior clearly has a negative impact on a variety of indices of
school adjustment. The research discussed in this chapter suggests that a
focus on the motivational processes that underlie children's aggression is
likely to be an effective intervention strategy for decreasing children's ag-
gressive behavior. Such decreases in aggression will then hopefully con-

tribute to more positive outcomes for children in school, including im-
proved academic performance and increased acceptance by classmates.
References
Asher, S. R., Parkhurst, J. T., Hymel, S., & Williams, G. A. (1990). Peer rejection
and loneliness in childhood. In S. R. Asher & J. D. Coie (Eds.), Peer rejection
in childhood (pp. 253-273). New York: Cambridge University Press.
Asher, S. R., & Renshaw, P. D. (1981). Children without friends: Social knowl-
edge and social-skill training. In S. R. Asher & J. M. Gottman (Eds.), The de-
velopment of children's friendships (pp. 273-296). New York: Cambridge Uni-
versity Press.
Bandura, A. (1979). Psychological mechanisms of aggression. In M. von
Cranach, K. Foppa, W. Lepenies, & D. Ploog (Eds.), Human ethology: Claims
and limits of a new discipline (pp. 316-379). Cambridge: Cambridge Universi-
ty Press.
Bandura, A. (1981). Self-referent thought: A developmental analysis of self-effi-
cacy. In J. H. Flavell & L. Ross (Eds.), Social cognitive development: Frontiers
and possible futures (pp. 200-239). Cambridge: Cambridge University Press.
Bandura, A. (1986). Social foundations of thought and action: A social cognitive
theory. Englewood Cliffs, NJ: Prentice-Hall.
Berkowitz, L. (1977). Situational and personal conditions governing reactions to
aggressive cues. In D. Magnusson & N. S. Endler (Eds.), Personality at the
crossroads: Current issues in interactional psychology (pp. 165-171) New
York: Wiley.
Berkowitz, L. (1993). Aggression: Its causes, consequences, and control. New
York: Academic Press.
Berndt, T. J. (1987, April). Changes in friendship and school adjustment after the
transition to junior high school. Paper presented at the biennial meeting of the
Society for Research in Child Development, Baltimore.
Bierman, K. L., Miller, C. L., & Stabb, S. D. (1987). Improving the social behav-
ior and peer acceptance of rejected boys: Effects of social skill training with in-
structions and prohibitions. Journal of Consulting and Clinical Psychology, 55,
194-200.
Block, J. H. (1983). Differential premises arising from differential socialization of
the sexes: Some conjectures. Child Development, 54, 1335-1354.
Coie, J. D., & Dodge, K. A., & Kupersmidt, J. B. (1990). Peer group behavior and
social status. In S. R. Asher & J. D. Coie (Eds.), Peer rejection in childhood (pp.
17-59). New York: Cambridge University Press.
Coie, J. D., & Koeppl, G. K. (1990). Adapting intervention to the problems of ag-
gressive and disruptive rejected children. In S. R. Asher & J. D. Coie (Eds.),
Peer rejection in childhood (pp. 309-337). New York: Cambridge University
Press.
Coie, J. D., & Krehbiel, G. (1984). Effects of academic tutoring on the social sta-
tus of low-achieving, socially rejected children. Child Development, 55,
1465-1478.
Crick, N. R., & Dodge, K. A. (1994). A review and reformulation of social infor-
mation-processing mechanisms in children's social adjustment. Psychological
Bulletin, 115, 74-101.
Crick, N. R., & Grotpeter, J. K. (1995). Relational aggression, gender, and social-
psychological adjustment. Child Development, 66, 710-722.
Crick, N. R., & Ladd, G. W. (1990). Children's perceptions of the outcomes of so-
612-620.
Dodge, K. A. (1980). Social cognition and children's aggressive behavior. Child
Dodge, K. A. (1986). A social information processing model of social compe-
tence in children. In M. Perlmutter (Ed.), The Minnesota Symposium on Child
Psychology (vol. 18, pp. 77-125). Hillsdale, NJ: Erlbaum.
Dodge, K. A., & Feldman, E. (1990). Issues in social cognition and sociometric
status. In S, R. Asher & J. D. Coie (Eds.), Peer rejection in childhood (pp.
119-155). New York: Cambridge University Press.
Dodge, K. A., & Frame, C. L. (1982). Social cognitive biases and deficits in ag-
gressive boys. Child Development, 53, 620-635.
Dodge, K. A., McClaskey, C. L., & Feldman, E. (1985). A situational approach to
the assessment of social competence in children. Journal of Consulting and
Clinical Psychology 53, 344-353.
Dodge, K. A., & Newman, J. P. (1981). Biased decision-making processing in ag-
gressive boys. Jo urnal of Abnormal Psychology, 90, 375-379.
Erdley, C. A., & Asher, S. R. (in press). Children's social goals and self-efficacy
perceptions as influences on their responses to ambiguous provocation. Child
Development.
Erdley, C. A., & Asher, S. R. (1993, August). Linkages between aggression and
children's legitimacy of aggression beliefs. In S. R. Asher (Chair), Social rela-
tionships, social beliefs, and aggression. Symposium conducted at the annual
meeting of the American Psychological Association, Toronto.
Fitzgerald, P. D., & Asher, S. R. (1987, August). Aggressive-rejected children s at-
tributional biases about liked and disliked peers. Paper presented at the annual
meeting of the American Psychological Association, New York.
Graham, S., Hudley, C, & Williams, E. (1992). Attributional and emotional deter-
minants of aggression among African-American and Latino young adolescents.
Graham, S., & Hudley, C. (1994). Attributions of aggressive and nonaggressive

African-American male early adolescents: A study of construct accessibility.
Guerra, N. G., & Slaby R. G. (1990). Cognitive mediators of aggression in adoles-
cent offenders: 2. Intervention. Developmental Psychology, 26, 269-277.
Hudley, C, & Graham, S. (1993). An attributional intervention to reduce peer-di-
rected aggression among African-American boys. Child Development, 64,
124-138.
Kupersmidt, J. B., Coie, J. D., & Dodge, K. A. (1990). The role of poor peer rela-
Press.
Ladd, G. W. (1981). Effectiveness of a social learning method for enhancing
children's social interaction and peer acceptance. Child Development, 52,
171-178.
Ladd, G. W., & Price, J. M. (1987). Predicting children's social and school adjust-
ment following the transition from preschool to kindergarten. Child Develop-
ment, 58, 1168-1189.
Nasby, W., Hayden, B., & DePaulo, B. M. (1980). Attributional bias among ag-
gressive boys to interpret unambiguous social stimuli as displays of hostility.
Journal of Abnormal Psychology, 89, 459—468.
Oden, S., & Asher, S. R. (1977). Coaching children in social skills for friendship
making. Child Development, 48, 495-506.
Parke, R. D., & Slaby R. G. (1983). The development of aggression. In E. M.
Hetherington (Ed.), P. H. Mussen (Series Ed.), Handbook of child psychology.
Vol. 4: Socialization, personality, and social development (pp.547-641). New
York: Wiley.
Perry, D. G., & Bussey, K. (1977). Self-reinforcement in high- and low-aggressive
boys following acts of aggression. Child Development, 48, 653-657.
Perry, D. G., Perry, L. C, & Rasmussen, P. (1986). Cognitive social learning me-
diators of aggression. Child Development, 57, 700-711.
Pervin, L. A. (1989). Goal concepts in personality and social psychology: A his-
torical perspective. In L. A. Pervin (Ed.), Goal concepts in personality and so-
cial psychology (pp. 1-17). Hillsdale, NJ: Erlbaum.
Pettit, G. S., Dodge, K. A., & Brown, M. M. (1988). Early family experience, so-
cial problem solving patterns, and children's social competence. Child Devel-
opment, 59, 107-120.
Rabiner, D. L., & Gordon, L. V. (1992). The coordination of conflicting social
goals: Differences between rejected and nonrejected boys. Child Development,

63, 1344-1350.
Renshaw, P. D., & Asher, S. R. (1982). Social competence and peer status: The
distinction between goals and strategies. In K. H. Rubin & H. S. Ross (Eds.),
Peer relationships and social skills in childhood (pp. 375—395). New York:
Springer-Verlag.
Renshaw, P. D., & Asher, S. R. (1983). Children's goals and strategies for social
interaction. Merrill-Palmer Quarterly, 29, 553-574.
Rockhill, C. M, & Asher, S. R. (1992, April). Peer assessment of the behavioral
characteristics of poorly accepted boys and girls. Paper presented at the annual
meeting of the American Educational Research Association, San Francisco.
Rubin, K. H., & Krasnor, L. R. (1986). Social-cognitive and social behavioral per-
spectives on problem solving. In M. Perlmutter (Ed.), The Minnesota Sympo-
sium on Child Psychology (vol. 18, pp. 1-68). Hillsdale, NJ: Erlbaum.
Slaby, R. G., & Guerra, N. G. (1988). Cognitive mediators of aggression in ado-
lescent offenders: 1. Assessment. Developmental Psychology, 24, 580-588.
Taylor, A. R., & Asher, S. R. (1984). Children's goals and social competence: In-
dividual differences in a game-playing context. In T. Field, J. L. Roopnarine, &
M. Segal (Eds.), Friendships in normal and handicapped children (pp.53-77).
Norwood, NJ: Ablex.
Wentzel, K. R. (1991). Social and academic goals at school: Motivation and
achievement in context. In M. Maehr & P. Pintrich (Eds.), Advances in motiva-
tion and achievement (vol. 7, pp. 185-212). Greenwich, CT: JAI Press.
Wentzel, K. R., & Erdley, C. A. (1993). Strategies for making friends: Relations to
social behavior and peer acceptance in early adolescence. Developmental Psy-
chology, 29, 819-826.
Zahavi, S., & Asher, S. R. (1978). The effect of verbal instructions on preschool
children's aggressive behavior. Journal of School Psychology, 16, 146-153.
Motivational opportunities and obstacles
associated with social responsibility and
caring behavior in school contexts
Martin E. Ford
The development of responsible and caring behavior patterns is a major

concern of parents and professionals responsible for the education and so-
cialization of adolescents (Damon, 1995). For example, in a recent study
(Krumboltz, Ford, Nichols, & Wentzel, 1987) in which parents and teach-
ers of several hundred high-school students were asked to rate 120 goal
statements in terms of "How important is it that people achieve this out-
come by age 18?" the top four goals were all concerned with aspects of
social responsibility and caring behavior:
1. Know right from wrong and act accordingly:

2. Be honest in dealing with other people:
3. Understand and avoid the harmful effects of prescription and ille-
gal drugs; and
4. Avoid stealing or damaging other people's property
A number of other social goals reflecting altruism and concern for others,
respect for people from diverse backgrounds, and responsible citizenship
were also rated very highly. Given the breadth and diversity of goal state-
ments included in this study - goal statements were constructed to repre-
sent five academic domains (verbal, math, science, social studies, and
fine arts) and five nonacademic domains (health, interpersonal compe-
tence, moral development, career development, and attitudes toward self
and learning) - this is compelling evidence for the strength and pervasive-
ness of society's concern with the development of social responsibility
and caring behavior in youth.
This broad developmental domain, which includes two conceptually
distinct but functionally interdependent sets of achievements - those asso-
ciated with social obligation (i.e., keeping commitments, fulfilling social
role expectations, conforming to social and moral rules, avoiding unethi-
126
Motivational opportunities and obstacles 111
cal or illegal conduct), and those associated with facilitation of others'

well being (i.e., providing others with validation, support, assistance, ad-
vice, etc.) - is not only a major concern in its own right. There is growing
evidence that the values and habits associated with social responsibility
and caring behavior are facilitative of learning and intellectual develop-
ment in school contexts (Wentzel, 1989, 1991a, 1991b, 1993a, 1993b, this
volume). Moreover, it is evident that the primary connection between the
social and academic domains is motivational in nature. Specifically, it ap-
pears that the goals, emotions, and personal agency beliefs that support
social responsibility and caring behavior provide a strong motivational
foundation for engagement and effort in school contexts. The main pur-
pose of this chapter is to outline the specific nature of these motivational
processes, and to underline the need for research and intervention work
focused on the development of these processes in school contexts.
The present volume represents an important effort to help meet this
need. However, this is just a beginning. The discrepancy between what we
know and what we need to know is embarrassingly large. Indeed, the field
is still struggling with the basic questions of how to frame the problem and
how to conceptualize the phenomena of interest (Wentzel, this volume).
Given the strength and persistence of society's concern with social respon-
sibility and caring behavior (Damon, 1995; Krumboltz et al., 1987), this is
a rather surprising state of affairs. Yet it is consistent with the historical em-
phasis in the adolescent literature (and the psychological literature in gen-
eral) on self-enhancing qualities such as autonomy, self-esteem, identity
development, and acceptance from peers (Batson, 1990; Feldman & Elliott,
1990). Moreover, when other-enhancing aspects of development have been
addressed, the results have been - with a few notable exceptions - of limit-
ed utility, in large part because motivational processes have been neglected.
For example, although the topic of moral development has received consid-
erable attention in the adolescent literature, the dominant guiding perspec-
tive (i.e., Kohlberg's stage theory moral development) has been one with a
peculiarly intellectual and individualistic focus, as a number of feminist
scholars have noted (e.g., Noddings, 1984, 1992). The relatively abundant
literature on prosocial development in young children offers some guid-
ance for adolescent and adult researchers, particularly with respect to in-
formation on the emotional regulation and socialization antecedents of
prosocial behavior (e.g., Eisenberg & Mussen, 1989; Radke-Yarrow, Zahn-
Waxler, & Chapman, 1983); nevertheless, that guidance is limited by the
very different cognitive, social, and motivational parameters involved in
these developmental periods.
With respect to intervention, the role of the schools in promoting the

development of social responsibility and caring behavior in youth can
arouse controversy when educators focus on lifestyle issues and personal
values (e.g., sexuality; moral and religious values). Nevertheless, there is
strong support for the involvement of schools in social problems that have
a clear behavioral focus and that produce significant public consequences
- for example, problems relating to citizenship, public health issues, and
respect for other's rights and property (Krumboltz et al., 1987). Thus, as
long as educators respect these boundaries, the question is not whether
schools should play a significant role in helping young people develop
into concerned, responsible adults; rather, the question is how schools can
best facilitate such development. This brings us back once again to the
critical need for more - and more sophisticated - research on the develop-
ment of social responsibility and caring behavior in youth. Such research
depends in turn on the availability of a theoretically sound and pragmati-
cally useful conceptual framework for understanding motivation and its
influence on competence development. One such conceptual framework,
Motivational Systems Theory, is outlined next.
Motivational systems theory

Motivational Systems Theory (MST; M. Ford, 1992) is a relatively new,
integrative theory of motivation derived from Donald Ford's Living Sys-
tems Framework (LSF; D. Ford, 1987; M. Ford & D. Ford, 1987). The LSF
is a comprehensive theory of human functioning and development that is
designed to represent all of the biological, psychological, and behavioral
processes of the person and how they are organized in complex patterns of
unitary functioning in variable environments. MST elaborates on a subset
of these processes, namely, those representing the three sets of phenomena
that have traditionally been of concern in the field of human motivation:
the selective direction of behavior patterns (i.e., what people want and
what they are trying to accomplish), the selective energization of behavior
patterns (i.e., how people get "turned on" or "turned off"), and the selec-
tive regulation of behavior patterns (i.e., how people decide to try some-
thing, stick with it, or give up.)
MST begins with the general idea that, in any given domain, effective
functioning or competence can best be defined in terms of the attainment
of personally and/or socially valued goals (M. Ford, 1985, 1992). MST
further proposes, using a heuristic (i.e., nonmathematical) equation, that
there are four basic prerequisites for goal attainment:
Competence = Motivation x Skill x R sive Environment

Biology
In other words, valued outcomes will not occur unless (1) the person has
the motivation needed to initiate and maintain activity directed toward
those outcomes; (2) the person has the skill needed to construct and exe-
cute a pattern of activity that is appropriate and effective with respect to
those outcomes; (3) the person's biological structure and functioning is
able to support both the motivational and skill components; and (4) there
is a responsive environment facilitating, or at least not excessively con-
straining, progress toward the desired outcomes (M. Ford, 1992). If any of
these components is missing or inadequate, ineffective functioning will
occur.
Because the biological and skill prerequisites for behaving in a respon-
sible or caring manner are well within the repertoires of most older chil-
dren and adolescents, the psychological literature on prosocial develop-
ment has focused largely on motivational and contextual influences on
social responsibility and caring behavior (e.g., Batson, 1990; Batson,
Dyck, Batson, Powell, McMaster, & Griffitt, 1988; Bergin, 1987; Chap-
man, Zahn-Waxier, Cooperman, & Iannotti, 1987; Eisenberg & Miller,
1987; Ford, Wentzel, Siesfeld, Wood, & Feldman, 1986; Ford, Wentzel,
Wood, Stevens, & Siesfeld, 1989; Hoffman, 1982; Krebs, 1975; Wentzel,
1994, this volume). This is not a well organized literature, however. Moti-
vational processes are usually studied in isolation rather than as compo-
nents of larger motivational patterns, and key concepts are often defined
differently by different researchers. For example, empathy has been a
dominant topic in this literature despite the fact that it is only one of many
relevant motivational processes, and theorists are still struggling with
some basic definitional issues. The social support literature also illustrates
how conceptual narrowness and uncertainty can constrain a productive
and exciting line or research. It thus seems likely that further progress will
depend on the development of conceptual approaches that are both more
precise and more comprehensive than existing frameworks.
In MST, motivation is understood to be a psychological, anticipatory,
and evaluative phenomenon (M. Ford, 1992). In other works, motivation
exists within (not outside) the person, where its job is to imagine possible
future states and outcomes, and to evaluate the extent to which those fu-
ture circumstances are desirable, emotionally compelling, and attainable.
Based on these defining criteria, motivation is defined in MST as the or-
ganized patterning of an individual's personal goals, emotional arousal
processes, and personal agency beliefs (M. Ford, 1992). This definition of
motivation can be represented symbolically by the following heuristic for-
mula:
Motivation = Goals x Emotions * Personal Agency Beliefs
Thus, for example, MST would predict that social responsibility and car-
ing behavior would be largely a function of (1) the degree to which per-
sonal goals are directed toward, or at least well aligned with, prosocial ac-
tions, and achievements; (2) the degree to which compelling emotions
inhibiting selfish or hurtful behavior, and facilitation caring and responsi-
ble behavior, are aroused in the pursuit of these goals; and (3) the degree
to which a person believes that their skills and surrounding context will
enable them to behave in a helpful or responsible manner.
Goals, emotions, and personal agency beliefs work together to guide
decision making - all are necessary, and none are sufficient for goal-di-
rected activity. Some compensation for weak elements in the equation is
possible, but each motivational component has "veto power" over the rest.
Strong, effective motivational messages to the "instrumental troops" (at-
tention, perception, memory, information processing, planning, problem-
solving, and action components of the person) will only result when there
is reasonable agreement in "motivational headquarters" that a goal is in-
deed desirable, compelling, and attainable (M. Ford, 1992). Conflicts be-
tween multiple, competing goals can also produce mixed motivational
messages and thereby reduce the effectiveness of ongoing behavior pat-
terns.
With this conceptual overview, it is now possible to explicate the func-
tional and substantive nature of each of these three sets of motivational
processes.
Personal goals
In MST, goals are defined simply as thoughts about desired (and unde-
sired) future states and outcomes (M. Ford, 1992). Such thoughts repre-
sent what the person is trying to accomplish (or avoid) in ongoing inter-
actions with the environment. The primary function of these thoughts is
to direct the other psychological and behavioral components of the per-
son to try to produce the desired consequences. Goals thus play a lead-
ership role in motivational patterns by defining their content and direc-
tion. They also serve as the central organizing force in social and
personality development (Cantor & Harlow, 1994; Emmons, 1989;

Markus & Ruvolo, 1989).
Goals that are clearly conceived and highly prioritized with respect to
potentially competing goals will organize and direct people's attention and
activity toward pursuit of those goals. For example, a student who values
teacher acceptance and conformity to classroom rules and expectations is
a good bet to behave in a socially responsible manner in school settings,
even when such behavior negatively impacts evaluations from peers
(Wentzel, 1994). Similarly, teenagers with a clear sense of what they are
trying to accomplish at school - whether those goals are academic, social,
or extracurricular - are much more likely to attend school and display pat-
terns of meaningful engagement than are those with little personal invest-
ment in the variety of goals afforded by the school context.
In MST, the Ford and Nichols Taxonomy of Human Goals (Ford, &
Nichols, 1987; 1991) is used to represent the diversity of goals that may
direct people's behavior in different situations (see Table 6.1) Although
people's thoughts about desired and undesired consequences are usually
rather personalized and context-specific, the taxonomy serves as a useful
heuristic device for researchers and practitioners who need a systematic,
standardized way of representing the basic "themes" in people's concerns
and priorities.
Theoretically, the goals most relevant to social responsibility and car-
ing behavior are the four integrative social relationship goals (see Table
6.1). In the case of belongingness, such behavior is an expected byproduct
of the individual's desire to enhance the functioning or integrity of a val-
ued social group (e.g., family, school, or peer group) or interpersonal rela-
tionship (e.g., friend, parent, teacher). In other words, acting in a responsi-
ble, caring way toward others helps create and cement social bonds.
Similarly, to the extent that participation and engagement in classroom
life is seen as a means to maintaining or enhancing these bonds (e.g., with
teachers or successful peers), belongingness can facilitate school achieve-
ment as well (Connell & Wellborn, 1991).
Social responsibility goals provide an alternative, more direct pathway
to responsible behavior by focusing attention on personal and/or cultural
standards of conduct, often independent of close relationships. Such goals
are reflected in efforts to keep interpersonal commitments, fulfill social
role obligations, and avoid social and ethical transgressions. These goals
appear to be an important part of the motivational foundation needed to
facilitate school adjustment and academic achievement (Ford, 1992;
Wentzel, 1991a, 1991b, 1993a, 1993b).
Table 6.1 The Ford and Nichols taxonomy of human goals

Desired Within-Person Consequences
Affective Goals -feelings or emotions one wishes to experience or avoid

Arousal Experiencing excitement or heightened arousal; Avoiding boredom
or stressful inactivity
Tranquility Feeling relaxed and at ease; Avoiding stressful overarousal
Happiness Experiencing feelings of joy, satisfaction, or well being; Avoiding
feelings of emotional distress or dissatisfaction
Bodily Sensations Experiencing pleasure associated with physical sensations, physical
movement, or bodily contact; Avoiding unpleasant or uncomfortable
bodily sensations
Physical Well Being Feeling healthy, energetic, or physically robust; Avoiding feelings
of lethargy, weakness, or ill health
Cognitive Goals - mental representations one wants to construct or maintain

Exploration Satisfying one's curiosity about personally meaningful events;
Avoiding a sense of being uninformed or not knowing what's going
on
Understanding Gaining knowledge or making sense out of something; Avoiding
misconceptions, erroneous beliefs, or feeling of confusion
Intellectual Creativity Engaging in activities involving original thinking or novel or
interesting ideas; Avoiding mindless or familiar ways of thinking
Positive Self-Evaluations Maintaining a sense of self-confidence, pride, or self-worth;
Avoiding feelings of failure, guilt, or incompetence
Subjective Organization Goals - special states that one seeks to experience or avoid
Unity Experiencing a profound or spiritual sense of connectedness,
harmony, or oneness with people, nature, or a greater power;
Avoiding feelings of psychological disunity or disorganization
Transcendence Experiencing optimal or extraordinary states of functioning;
Avoiding feeling trapped within the boundaries ordinary
experience
Desired Person- Environment Consequences
Self-Assertive Social Relationship Goals - maintaining or promoting the self

Individuality Feeling unique, special, or different; Avoiding similarity or
conformity with others
Self-Determination Experiencing a sense of freedom to act or make choices; Avoiding
the feeling of being pressured, constrained, or coerced
Superiority Comparing favorably to others in terms of winning, status, or
success; Avoiding unfavorable comparisons with others
Resource Acquisition Obtaining approval, support, assistance, advice, or validation from
others; Avoiding social disapproval or rejection
Table 6.1 Continued

Desired Person- Environment Consequences
Integrative Social Relationship Goals - maintaining or promoting the well being of others
Belongingness Building or maintaining attachments, friendships, intimacy, or a
sense of community; Avoiding feelings of social isolation or
separateness
Social Responsibility Keeping interpersonal commitments, meeting social role
obligations, and conforming to social and moral rules; Avoiding
social transgressions and unethical or illegal conduct
Equity Promoting fairness, justice, reciprocity, or equality; Avoiding unfair
or unjust actions
Resource Provision Giving approval, support, assistance, advice, or validation to others;
Avoiding selfish or uncaring behavior
Task Goals -producing desired effects or avoiding undesired effects in the relationship between
the person and environment
Mastery Meeting a challenging standard of achievement or improvement;
Avoiding incompetence, mediocrity, or decrements in performance
Task Creativity Engaging in activities involving artistic expression or creativity;
Avoiding tasks that do not provide opportunities for creative action
Management Maintaining order, organization, or productivity in daily life tasks;
Avoiding sloppiness, inefficiency, or disorganization
Material Gain Increasing the amount of money or tangible goods one has;
Avoiding the loss of money or material possessions
Safety Being unharmed, physically secure, and free from risk; Avoiding
threatening, depriving or harmful circumstances
Equity goals also promote social responsibility and caring behavior by

motivating people to try to change circumstances characterized by social
injustice, unequal resource distribution, and the victimization of helpless
or disadvantaged individuals. Like social responsibility goals, such con-
cerns may be manifested even in the absence of close relationships. Equi-
ty goals have not been directly linked to academic outcomes; however, it
may be a question worthy of some exploration, given the demonstrated
importance of other kinds of integrative social relationship goals to school
achievement.
Finally, resource provision, with its focus on the needs and welfare of
others, is the most directly implicated category of goals with respect to
caring behavior. Although such goals sometimes emerge in interactions
with strangers and acquaintances, they are usually embedded in social re-
lationships involving reciprocal exchange processes (e.g., friendship or
dating relationships) or in asymmetrical social roles in which one person
is responsible for providing resources to another (e.g., parent-child or

teacher-student relationships). Resource provision goals may also be an
important feature of effective cooperative learning arrangements, perhaps
in conjunction with social responsibility and resource acquisition con-
cerns (Slavin, 1981, 1987).
Personal agency beliefs

This motivational component refers to whether a person believes that the
other fundamental components of effective person-in-context functioning
(i.e., the skill and "responsive environment" components) can be counted
on to facilitate progress toward a goal. Thus, there are two different types
of personal agency beliefs; capability beliefs (i.e., expectancies about
whether one has the personal skill needed to attain a goal), and context be-
liefs (i.e., expectancies about whether the environment will be supportive
of efforts to attain a goal) (M. Ford, 1992). Presumably, the personal
agency beliefs of greatest significance for social responsibility and caring
behavior are those pertaining to integrative social relationship goals. For
example, one would expect students who believe they are capable of ad-
justing to the interpersonal expectations of the classroom, and who are
confident they will be supported in their efforts to do so to be more social-
ly and academically successful in that context than those with significant
doubts about either of these prerequisites for effective functioning.
The function of personal agency beliefs is to help people decide which
goals to pursue and how much time and effort they should invest in pur-
suit of a particular goal. This implies that, if a person is disinterested in a
particular goal, his or her capability and context beliefs with respect to
that goal will be of no significance. For example, if one cares little about
mathematics, beliefs about the possible consequences of trying to achieve
in mathematics will be moot (Meece, Wigfield, & Eccles, 1990). Con-
versely, personal agency beliefs can be expected to play a very influential
role in the activation and inhibition of behavior when there is a strong goal
in place. For example, people often do not act on deeply felt concerns
about another person's welfare because they doubt their ability to act ef-
fectively or are concerned about how their actions will be interpreted.
Along these lines, it is important to keep in mind that motivation itself
is only one of four major elements contributing to effective functioning.
One should not assume, therefore, that strong goals and positive capabili-
ty and context beliefs are sufficient for goal attainment. Ultimately, peo-
ple must also have real skills and a truly responsive environment. Never-
theless, personal agency beliefs are often more fundamental than the actu-
al skills and circumstances they represent in the sense that they can moti-
vate people to create opportunities and acquire capabilities they do not yet
possess. Thus, like personal goals, personal agency beliefs serve a crucial
developmental function through their role in creating and maintaining pat-
terns of social and academic engagement and achievement.
Emotional arousal processes

Emotions are organized functional patterns consisting of three compo-
nents: an affective (neural-psychological) component (i.e., the general
subjective feeling part of the emotion), & physiological component (i.e., a
supporting pattern of biological processing), and a transactional compo-
nent (i.e., expressive gestures that influence relevant aspects of the con-
text) (D. Ford, 1987). Thus, emotions help people deal with varying cir-
cumstances by providing evaluative information about the person's
interactions with the environment (affective regulatory information) and
by supporting and facilitating action designed to produce desired conse-
quences (energizing function).
Emotions provide a very potent mechanism for regulating behavior be-
cause affective experience has an immediacy to it that is hard to ignore
(Frijda, 1988). For example, when people experience intense feelings of
guilt or compassion, it is difficult to attend to anything else except the
concern associated with those feelings (e.g., "I must make amends" or
"Something needs to be done to help this person"). Emotions also gain
potency by virtue of their flexibility - by interpreting situations in differ-
ent ways, people can learn to associate particular emotions with almost
any conceivable context (D. Ford, 1987). For example, students can learn
to feel anxious in evaluative classroom situations, even when the objective
consequences of a negative evaluation are imagined or trivial. Moreover,
once learned, emotional associations may be generalized to other related
contexts, including those that the individual has not yet encountered (e.g.,
other performance-related social settings).
Emotions and personal goals are highly interdependent processes. In-
deed, Frijda (1988) has postulated what he calls the "Law of Concern" to
characterize this interdependency: "Emotions arise in response to events
that are important to the individual's goals, motives, or concerns" (p. 351).
Emotions are also closely linked to personal agency beliefs, with both
serving to regulate (activate or inhibit) behavior in specialized ways. Per-
sonal agency beliefs, because they are more likely to incorporate consid-
erations about long-term consequences and the potential availability of al-

ternative courses of action, are particularly useful for decisions about "big
picture" matters, such as whether one should initiate or continue a com-
plex, difficult, or time-consuming course of action. Emotions, on the oth-
er hand, are particularly useful when effective functioning requires imme-
diate or vigorous action in the context of a concrete problem or
opportunity, such as escape from imminent harm, removal of an obstacle
to goal attainment, or inhibition of a personally or socially damaging ac-
tion (M. Ford, 1992).
Because emotions are so flexible, social responsibility and caring be-
havior may be facilitated by a diversity of emotion patterns, both positive
and negative. However, the research literature on the emotional regulation
of prosocial behavior suggests that there are three emotion patterns of par-
ticular significance: empathic concern, guilt, and pride, in helping others
(e.g., Batson et al., 1988; Eisenberg & Miller, 1987; Ford et al., 1989;
Hoffman, 1982).
An empirical study of caring competence

In this section, the results of a recent investigation of the motivational
processes associated with caring behavior in youth (Ford, Love, Chase,
Pollina, & Ito, 1991) are summarized in an effort to illustrate the develop-
mental significance of social motivation in this domain of development.
Where appropriate, connections are drawn between this research and re-
lated work on adolescent social responsibility.
Participants in this study were 174 students (99 girls and 75 boys) en-
rolled in an ethnically diverse California (Bay Area) public high school lo-
cated in an upper-middle class community. The study was designed to test
the hypothesis that the three most fundamental qualities of caring behav-
ior patterns are: (1) strong integrative goals (especially resource provi-
sion); (2) powerful emotions facilitating prosocial behavior and inhibiting
selfish behavior (especially empathetic concern); and (3) positive person-
al agency beliefs for behaving in a caring manner. Caring competence was
defined in this study as one aspect of intergrative competence, a broad
concept representing effective efforts to promote, protect, maintain, or re-
store the rights, interest, or well being of another person or group of peo-
ple. Within this domain, caring competence refers specifically to situa-
tions in which prosocial action is more a matter of personal choice than it
is a function of social obligation. In other words, caring competence is
most clearly manifested in situations for which there are no specific laws,
social rules, or interpersonal commitments requiring one to behave in a

prosocial manner. The concept of social responsibility represents these
more obligatory aspects of integrative competence.
Caring competence
The concept of caring competence was measured in this study by seeking
information from a diversity of sources. Specifically, self-ratings, teacher
ratings, and peer nominations of caring competence were obtained with
respect to a variety of concrete, hypothetical situations emphasizing the
possibility of behaving in a helpful, supportive, or civic-minded manner.
Situations were designed so that external pressures and obligations to be-
have in a caring manner were minimized. In each case, the caring re-
sponse option required the student to behave in a socially exceptional way,
either in terms of "going beyond the call of duty," resisting pressure to
conform to peer norms, or overcoming the temptation to behave in a per-
sonally expedient manner.
The primary tool used to assess caring competence was the Caring
Competence Nomination Form (CCNF), a six-item measure adapted from
a similar instrument developed by M. Ford (1982) to assess self-assertive
and integrative social competence (broadly defined). In the first part of
the CCNF, students were asked to nominate up to three boys and three
girls in their class (other than themselves) who would be particularly com-
petent in a diversity of situations requiring some form of caring behavior
(e.g., being a tutor or peer counselor; doing community volunteer work;
supporting a teacher or student in distress). Students were then asked to
respond to two questions regarding their own level of caring competence
in each situation. Teachers were also asked to rate how competent each of
the students in their class(es) would be in each of the six situations in the
CCNF.
In addition to the CCNF, students completed a revised version of the
Youth Decision-Making Questionnaire (YDMQ), an eight-item measure
based on a similar questionnaire used by Ford et al. (1989) in their study
of adolescent social responsibility. The revision involved the insertion of
new situations into the same format used in the original questionnaire.
Specifically, students were asked to indicate on a four-point scale whether
they would "Definitely" or "Probably" choose to behave in a caring or
noncaring response option.
As expected, correlations among the various measures of caring com-
petence were all positive and significant, ranging from .24 to .61. This
made it possible to represent this concept using a single composite vari-

able that would presumably yield a more valid representation of students'
level of caring competence than could be provided by any of the assess-
ments taken separately.
Personal goals
Two measures were used in this study to assess the general importance of
integrative goals in students' lives. The first was the Assessment of Per-
sonal Goals (APG), a self-administered paper-and-pencil measure de-
signed to assess the strength of each of the 24 categories represented in
the Ford and Nichols Taxonomy of Human Goals (see Table 6.1). Al-
though only four of these scales (i.e., the Resource Provision, Equity, So-
cial Responsibility, and Belongingness scales) were directly relevant to
the hypotheses guiding the study, the entire measure was used to explore
the possibility that caring competence might be associated with other goal
categories. For example, the use of the full APG made it possible to ex-
plore the possibility that caring competence might be negatively associat-
ed with various kinds of self-enhancing goals.
The APG is composed of 24 five-item scales, with each item presented
in situation-specific form. For each item, respondents are asked to indi-
cate on a 9-point scale how likely they would be to desire a particular out-
come or to be bothered by a failure to attain that outcome (e.g., "A friend
of yours is moving across the country to take a new job. Would it be im-
portant to you to stay in close touch with this friend by mail or phone?").
The second measure used to assess personal goals, the Assessment of
Personal Agency Beliefs (APAB), is also based on the Ford and Nichols
Taxonomy of Human Goals. Although its primary purpose is to assess ca-
pability and context beliefs, respondents must first indicate for each do-
main, "How often is this an important goal for you in your everyday life?"
The inclusion of this preliminary question is based on the theoretical
premise that capability and context beliefs are uninterpretable if the goal
to which they refer is unimportant to the individual.
The version of the APAB used in this study asked students to respond
to 21 "Life Goals" (it has since been expanded to include all of the 24
goals included in the Ford and Nichols Taxonomy), including items repre-
senting all four intergrative goals: "Caring For and About Other People"
(Resource Provision), "Working For a More Just Society), and "Having
Close Personal Relationships" (Belongingness). As with the APG, the en-
tire APAB was administered (rather than just those items referring to inte-
grative goals) to explore the possibility that caring competence might be

associated with other goal categories. This also made it possible to look
for converging patterns of evidence across two somewhat different goal
measures.

After rating the importance of each goal in the APAB, students were in-
structed to respond to two additional questions representing their capabil-
ity and context beliefs, respectively, for attaining those goals: "Consider-
ing only YOURSELF - your skills, interests, and personality - do you
think you are capable of being successful in this area?" and "Considering
only things OUTSIDE of yourself- your family and friends, your school
or job situation, etc. - do you think they generally make it easier or harder
for you to succeed in this area?" (In a subsequent revision of the APAB,
this latter question was disaggregated into a set of separate questions, one
for each of the respondent's major life contexts).
Because the concept of perceived social support overlaps considerably
with the concept of context beliefs (i.e., social support is a major compo-
nent of environmental responsiveness), students were also asked to com-
plete the Youth Social Support Questionnaire (YSSQ), a brief assessment
designed to measure both the size of the student's social support network
and the degree to which the student felt satisfied with their (perceived)
level of social support. Students were asked to list the first name or initials
of each person that they felt close to, or to whom they could go for help,
encouragement, or advice, and then indicate how helpful each person on
this list would be if the student actually sought out their support (i.e.,
somewhat, fairly, or extremely helpful). Finally, students were asked to in-
dicate the extent to which they were satisfied with the size and helpfulness
of their social support network.
Emotional arousal processes

In addition to providing information about self-reported caring behavior,
the Youth Decision-Making Questionnaire also served as an assessment of
students' anticipated emotional responses to a diversity of caring-relevant
situations. Specifically, after responding to the behavioral choice question
in each of the eight situations in the YDMQ, students were asked to indi-
cate the extent to which they would anticipate feeling several different
emotions (i.e., very much, a little, or not at all) if they were to choose the
caring response option or the noncaring alternative (students responded to

both possibilities in each situation). The specific number and wording of
each category of emotion items varied somewhat across situations in order
to fit the particular circumstances involved; however, in each situation
there was at least one question pertaining to the emotions of empathic
concern, guilt, pride/pleasure in helping others, self-focused positive emo-
tions (e.g., relief at avoiding an uncomfortable situation), and self-focused
negative emotions (e.g., fear of peer disapproval).
Hypothesis 1: Integrative goals and caring competence. Table 6.2

summarizes the results pertaining to the importance of integrative goals as
qualities of caring adolescents. As expected, both the APG and APAB ver-
sions of each goal were significantly correlated with caring competence.
Moreover, the two highest correlations involved resource provision goals.
Inspection of the results for the component indices of caring competence
indicated that integrative goals were highly predictive of students' deci-
sion making with respect to caring behavior, and moderately informative
with regard to the quality of their performance in a caring role.
Hypothesis 2: Personal agency beliefs for integrative goals and caring

competence. Tests of the proposition that caring adolescents are likely
to have positive capability and context beliefs for the attainment of inte-
grative goals are also presented in Table 6.2. As predicted, personal
agency beliefs for integrative goals were reliably associated with caring
competence (with the exception of context beliefs in the social responsi-
bility domain). Unexpectedly, results were generally stronger for the capa-
bility belief variables than for the context belief measures. Although this
finding may be due in part to measurement problems (recall that the
APAB was later revised to enable assessment of context beliefs in differ-
ent contexts), it is consistent with the profile shown by a number of the
most caring students in the sample in which they saw themselves as strug-
gling to act on others' behalf in a relatively uncaring world.
Hypothesis 3: Emotional arousal processes and caring competence.

Correlations between the five emotion variables and the various indices of
caring competence are summarized in the lower half of Table 6.2. Consis-
tent with the literature's emphasis on empathy as an antecedent of caring
behavior, empathic concern was a powerful predictor of behavioral deci-
sion making in caring situations, and a moderately good indicator of how
effectively a person would be likely to perform in a caring role. Guilt and
Table 6.2 Correlations between motivational processes and caring

competence
Correlation of caring competence with:
Integrative Goals Capability Beliefs Context Beliefs Emotions
(APG/APAB) (APAB) (APAB) (YDMQ)
Resource provision/caring .54**/.49** .36** .15*
Equity/more just society .40**/.42** .35** .18*
Social responsibility .42**/.32** .14* -.12
Belongingness/close .35**7.17* .28** .14*
relationships
Level of social support .27**
(number of people)
Social support satisfaction .20*
Empathic concern .63**
Pride .47**
Guilt .53**
Positive self-focused emotions -.45**
Negative self- focused emotions .18*
N = 157-174.
*p < .05.
**/?<. 001.
APG = Assessment of Personal Goals; APAB = Assessment of Personal Agency Beliefs;
YDMQ = Youth Decision-Making Questionnaire
pride/pleasure in helping others manifested a similar pattern of correla-

tions, but at a slightly lower magnitude for most of the caring competence
criteria. Self-focused positive emotions (i.e., good feelings associated
with selfish or uncaring choices) were also reliable predictors of caring
behavior, but in the opposite direction. Self-focused negative emotions
(i.e., bad feelings associated with sacrificing selfish pleasures or being
unable to escape others' distress) repeated this pattern but at a much lower
level.
This same set of emotions appears to be involved in motivational pat-
terns associated with adolescent social responsibility (Ford et al., 1986;
Ford et al., 1989). However, in these studies guilt emerged as the most im-
portant emotion regulating this domain of social behavior, especially in
the absence of external sanctions for irresponsible behavior. This was par-
ticularly true for teenage boys, who were consistently less likely than girls
to report feelings of empathic concern in situations involving negative
consequences to others. The Ford et al. (1989) study also demonstrates the
additive effect of combining multiple emotions supportive of responsible
behavior, a phenomenon that appears to hold for caring behavior as well
(see Table 6.3). Evidently, it is particularly difficult to ignore emotional
Table 6.3. Comparison of students high and low in caring competence on

selected demographic and motivational variables
Probability (from 0 to 1) that high caring (N=25) and low caring (N=21) students will:
High caring Low caring
Demographic variables
Be a female .75 .33
Be engaged in some form of volunteer work or school .50 .19
or community service
Report having an A or A+ GPA .33 .24
Report having a GPA of C+ or lower .00 .24
Intergrative goals
Say that "Caring for and about others" (APAB) is "always" 1.00 .43
or "almost always" important
Resource provision .63 .00
Equity .50 .20
Social responsibility .25 .05
Belongingness .83 .25
Score low (1-5) on the APG goals of:
Resource provision .08 .55
Equity .04 .30
Social responsibility .08 .70

Say that they are "always" or "almost always" capable of .83 .23
caring for and about others
Say that they are "always" or "almost always" capable of .75 .14
working for a more just society
Score low (1 or 2 on a 5-point scale) on social support satisfaction .04 .25
Emotions
Score high (2.5-3) on the emotions of:
Empathic concern .88 .00
Pride/pleasure in helping others .58 .05
Guilt .29 .05
Score low (0-2) on the emotions of
Empathic concern .00 .33
Pride/pleasuere in helping others .00 .52
Guilt .00 .76
Score low (0-2) on both variables representing self-focused emotions .63 .14
Other goals associated with caring competence
Score high (7-9) on theAPG/APAB goals of:
Intellectual or Task creativity (APG) .83 .15
Unity (APG) or Spiritually (APAB) .67 .19
Positive self-evaluations or Individuality (APG) .75 .30
messages when there are two or more messengers providing guidance and
pressure to act in a particular way.
Gender differences in motivation and behavior associated with

caring competence
The female participants in this study appeared to be no more capable of
caring behavior than their male counterparts; however, they were signifi-
cantly more likely to choose to behave in a caring manner in situations
calling for such behavior. Consistent with this finding, significant gender
differences favoring females were observed on three of the four APG inte-
grative goals - Resource Provision, Social Responsibility, and Belonging-
ness - as well as the "Caring" APAB goal item. Alternatively, boys and
girls had similar scores on the APG and APAB equity measures, thus pro-
viding indirect support for those who distinguish between a "justice" and
"caring" orientation to morality (e.g., Noddings, 1984, 1992). Along these
same lines, females had significantly higher capability and context beliefs
for the goals of resource provision and belongingness, slightly more posi-
tive personal agency beliefs in the domain of social responsibility, and vir-
tually identical capability and context beliefs with regard to equity. The
girls in this study also reported having a somewhat larger social support
network (on average) than did males, although no reliable gender differ-
ence in social support satisfaction was observed.
An examination of gender differences on the emotion variables revealed
that, compared to males, females were significantly more likely to antici-
pate feeling empathic concern and positive emotions (pride/pleasure in
helping) in situations where they chose to try to help someone in need or in
distress. Females were also more likely to anticipate feeling empathic con-
cern and guilt when they chose not to provide assistance or support in such
situations. On the other hand, males were more likely to anticipate feeling
self-focused emotions focused on the positive qualities of the noncaring al-
ternative (e.g., relief at avoiding a socially awkward situation; excitement
about engaging in a fun experience that might otherwise be sacrificed).
These results precisely match those obtain by Ford et al. (1989) in their
study of emotional influences on adolescent social responsibility.
Comparison of high caring and low caring students

In an effort to determine whether the qualities associated with caring be-
havior as revealed by correlational analyses involving the whole sample
were truly characteristic of highly caring individuals, a. further set of

analyses was conducted focusing on students scoring particularly high or
low in caring competence. To insure that these "exemplars" of caring and
noncaring behavior were unambiguously prototypical of the groups they
were intended to represent, a more stringent criterion was applied than
simply scoring high or low on the caring composite variable. Specifically,
high and low caring students were required to have scores at least two-
thirds of a standard deviation above or below the mean (respectively) on
at least three of the four variables used to create the caring composite,
with the remaining score also in the same direction relative to the mean
for that variable. This procedure eliminated several students who might
have otherwise been selected on the strength of very high or very low
scores on a subset of measures (e.g., students with highly favorable self-
evaluations but less than flattering reputations among their teachers or
peers; students rated highly by their teachers but rarely nominated by their
classmates; students with consistently high scores on the CCNF but a rel-
atively uncaring pattern of decision making on the YDMQ). Nevertheless,
because of the overall consistency among the various indices of caring
competence for those scoring toward the tails of each distribution, 28% of
the sample met the criteria for membership in one of the extreme groups
(i.e., 24 students in the high caring group and 21 students in the low car-
ing group).
The results presented in Table 6.3 document the fact that the qualities
most closely associated with caring competence in the group-level analy-
ses were indeed consistently manifested in highly caring individuals. They
also reveal the remarkable extent to which the variables representing these
qualities discriminate between high and low caring students. In fact, a dis-
criminant function constructed from a subset of only 14 of these variables
(i.e., those most relevant to the study hypotheses) was sufficient to
achieve a 100% hit rate in a classification analysis designed to predict the
group membership of each student.
Among the specific findings listed in Table 6.3, several are particularly
noteworthy. As one might expect from the consistent gender differences
that emerged for various motivational and behavioral manifestations of
caring competence, girls were three times more likely to be in the high
caring group, and boys were twice as likely to be in the low caring group.
The strong association between caring behavior and academic achieve-
ment, which was not readily apparent from analyses testing for linear rela-
tionships (correlations between caring competence and students' self-re-
ported GPA ranged from .08 to .18), also appears to be worthy of further
research. Inspection of the data for individual students confirmed that

both the high and low ends of the caring competence continuum were well
represented, even among the academically superior students in the sample.
High achievement is clearly no guarantee that a student will be regarded
as a caring person. However, not a single student with a GPA of C+ or
lower was among those identified as part of a group of 24 caring "exem-
plars." In other words, although some high achieving students were identi-
fied as uncaring individuals, none of the caring exemplars in the study
was a low achiever. Wentzel has observed similar results with respect to
the relationship between social responsibility and academic achievement
(Wentzel, 1989, 1991a, 1993a, this volume). One might speculate that the
primary role of integrative social motivation in academic achievement is
to (a) strengthen classroom motivation through the additive impact of
multiple goals (M. Ford, 1992; Wentzel, 1989), and (b) provide some "mo-
tivational insurance" (M. Ford, 1992) in cases where other goals (e.g., ex-
ploration, understanding, mastery, intellectual or task creativity) are unre-
liable or absent.
Examination of the psychological variables represented in Table 6.3
suggests that the "core" of the caring adolescent is an easily activated mo-
tivational pattern characterized by strong resource provision goals, com-
pelling feelings of empathic concern and pride or pleasure in helping oth-
ers, and positive capability beliefs for caring action. Circumstances that
activate this motivational pattern are likely to afford the simultaneous at-
tainment of multiple goals that are personally compelling and rich in
meaning for highly caring people.
As expected, most of the students in the high caring group had elevated
scores on measures of belongingness, resource provision, and empathic
concern. Moreover, virtually no one in this group had low scores on any of
the goals or emotions reflecting concern for others. Surprisingly, though,
these students were also relatively unlikely to score high on indices of so-
cial responsibility and guilt. In contrast, a large majority of the low caring
students had unusually low scores on these measures - in some cases so
low as to be frightening. This suggests that a second motivational pattern
- one that is related to but qualitatively distinct from the basic caring mo-
tivational pattern - must be taken into account if one is to understand the
full range of individual differences in caring competence. Specifically, it
appears that a "backup" motivational pattern based on obligation may be
needed to inhibit those with little intrinsic interest in caring from behaving
in uncaring ways. In other words, some degree of desire to avoid feelings
of guilt and other consequences of socially irresponsible behavior (e.g.,
parental disapproval or peer rejection) may be necessary to produce mini-

mally acceptable levels of caring competence. On the other hand, this
obligation pattern would rarely be sufficient to produce high levels of car-
ing behavior - it promotes "doing one's duty" rather than "going beyond
the call of duty." It is nevertheless a profoundly important component of
social motivation, as it provides "motivational insurance" against danger-
ously low levels of social responsibility and caring behavior (M. Ford,
1992).
The role of social responsibility motivation in insuring socially accept-
able levels of caring competence appears to be analogous to its role in
helping students maintain satisfactory levels of academic effort and
achievement (Wentzel, 1989, 1991a, 1993a, this volume). This suggests
that, developmentally, social responsibility motivation may be an essential
prerequisite for both academic and social success. This in turn implies that
the development of strong goals, emotions, and personal agency beliefs
associated with social responsibility should be a major focus of parenting
and education in early childhood (Maccoby, 1992).
Individual motivational profiles were also examined in an effort to dis-
cern whether different "types" of high and low caring students could be
identified within each group. At least with respect to caring competence,
highly caring students seemed to be a relatively homogeneous group.
They uniformly displayed most of the features associated with the proto-
typical caring pattern at a very high level, and generally manifested the
qualities associated with the obligation pattern at a somewhat lower level
- that is, caring appeared to be something they wanted to do and liked to
do, not just something that they felt they had to do or ought to do.
Low caring students, on the other hand, appeared to comprise a more
heterogeneous group. They were generally distinguished by their very low
scores on resource provision goals and empathic concern. However, some
of these students did manifest a moderate or even high degree of concern
for others on certain variables. In such cases there were typically a number
of very prominent personal goals that seemed to overpower the integrative
concerns in the student's goal hierarchy. In particular, three different self-
focused personality "types" were evident at the level of the individual
low-caring student, with some students showing elements of more than
one type.
One subset of cases might best be described as representing a "fun" or
"feel good" type characterized by strong self-focused emotions and very
high scores on the goals of Entertainment, Physical Weil-Being, and Bod-
ily Sensations. In the context of relatively weak goals and emotions re-
fleeting concern for others, this hedonistic pattern presumably takes on a

rather self-centered or even self-absorbed quality. A second subset of low-
caring students seemed to manifest a "hypermasculine" pattern character-
ized by very high scores on goals such as Self-determination, Superiority,
and Material Gain. This pattern was coupled with low scores in an unusu-
ally large number of goal categories, thus implying a rather narrow focus
on goals representing power, control, and personal achievement. A third
type of pattern found in some low-caring students was a "pragmatic-de-
fensive" pattern focused primarily on the removal or avoidance of discrep-
ancies in one's immediate affective and perceptual experience. Goals re-
flecting a need for safety, emotional stability, and order were prominent in
the goals hierarchies of these students.
Finally, the results summarized in Table 6.3 suggest that, compared to
their less caring peers, highly caring adolescents are more likely to be in-
terested in and concerned about a diversity of personal goals - social and
nonsocial. Indeed, one of the distinguishing features of many students
scoring low in caring competence was the relatively large number of goals
they rated as unimportant. Endorsement of a broad range of goals did not
necessarily mean that a student would also score high in caring compe-
tence; however, highly caring students virtually never manifested the nar-
rowness characterizing the goal hierarchies of many of those scoring low
in caring competence.
Related to this pattern is what appears to be a strong tendency for high-
ly caring youth to be particularly interested in goals pertaining to intrinsi-
cally meaningful aspects of human experience - for example, creative ex-
periences (Intellectual and Task Creativity); experiences reflecting
feelings of unity or spiritual connectedness with people, nature, or a
greater power (Unity/Spirituality); and, to a lesser extent, experiences re-
lated to understanding and task accomplishment (Understanding, Mas-
tery/Skill Development, and Management/Personal Productivity). Consis-
tent with this interpretation, caring competence was negatively related to
Material Gain, a goal commonly regarded as a more extrinsic and superfi-
cial source of motivation. Thus, caring for others may be, in many cases, a
part of a larger pattern of seeking to engage life in ways that are particu-
larly or even profoundly rich in meaning. The strong (and unexpected)
positive relationship found between caring competence and the goal of In-
dividuality, which reflects a desire to feel unique, special, or different
from others, may also be a part of this pattern of being attracted to experi-
ences that carry enduring value and meaning (in contrast to experiences
that are more transient and superficial in meaning).
Caring competence was also positively and significantly related (albeit

at a somewhat lower level) to a set of goals pertaining to aspects of self-
acceptance and mental, emotional, and physical well being (i.e., Positive
Self-Evaluations/Self-Esteem, Happiness/Feeling Happy, Resource Ac-
quisition/Peer Acceptance, and Physical Well-Being/Physical Fitness).
Thus, it does not appear to be the case that investment in self-enhancing
goals is a major factor inhibiting caring behavior. Indeed, the present re-
sults support the opposite view, namely, that people who "look after them-
selves" (i.e., who maintain an adequate level of self-comfort and self-ac-
ceptance) are more likely to behave in a caring manner than those who fail
to do so. It may also be that for many people one of the primary rewards
of helping and supporting others is increased self-acceptance and a more
positive self-image.
Implications for educational research and practice

The research described in this chapter, despite some methodological limi-
tations (e.g., use of convenience samples and relatively new measures),
provides initial support for the utility of Motivational Systems Theory in
guiding research on adolescent social responsibility and caring behavior.
Indeed, it is hard to imagine how one could represent the multitude of mo-
tivational and nonmotivational factors involved in the development of so-
cial and academic competence without some sort of a complex systems
approach (D. Ford, 1987; M. Ford, 1992).
There appear to be numerous promising directions for further research
on adolescent social responsibility and caring behavior. Of particular in-
terest are the rich goal hierarchies of highly caring adolescents. The link
between social motivation and caring behavior appears to involve much
more than a few goals and emotions of straightforward relevance to proso-
cial behavior. Rather, strong patterns of caring behavior appear to be sup-
ported by a complex web of mutually reinforcing goals that make such be-
havior compelling and rewarding in many different ways. This is
consistent with the MST premise that strong motivational patterns will
typically be anchored by more than one goal (i.e., the Multiple Goals prin-
ciple; M. Ford, 1992). Application of a multiple goals perspective also
provides a way of thinking about how contexts might be designed to facil-
itate social responsibility and caring behavior despite the vast individual
differences in social motivation evident in a heterogeneous sample of ado-
lescents.
Further research on the reciprocal relationships between social compe-
tence and academic achievement may also profit from the application of a
multiple goals perspective (e.g., Wentzel, 1989, 1991a, 1993b, this vol-
ume). Responsible, caring teenagers consistently affirm (or at least do not
deny) the importance of goals related to learning, mastery, task comple-
tion, and intellectual and task creativity. This appears to be part of a more
general motivational pattern involving engagement with diverse aspects of
the world, high standards for personal conduct, and a strong desire to meet
those self-evaluative standards. Conversely, people with narrow interests
and weak internal standards appear to be particularly prone to expedient,
self-serving behavior (Bandura, 1991).
The paramount importance of social motivation in the development of
social and academic competence is thus becoming increasingly clear. If
one can facilitate the self-construction of a strong motivational foundation
with respect to the goals, emotions, and personal agency beliefs associat-
ed with social responsibility and caring behavior, and reinforce that foun-
dation on an everyday basis in schools, families, and other important life
contexts, positive outcomes will almost surely follow. Conversely, without
such a foundation, a child is very likely to experience negative social and
academic outcomes, and will be poorly prepared for adult responsibilities.
This emphasis on motivation as the psychological foundation for com-
petence development (M. Ford, 1992) leads to the question, "How can ed-
ucators best facilitate the development of the motivational qualities asso-
ciated with social responsibility and caring behavior?" The research
described in this chapter helps clarify the nature of the task to be accom-
plished, and suggests potential targets of intervention for programs, activ-
ities, and strategies designed to facilitate such behavior. For example, ex-
periences that focus on very specific resource provision problems and
objectives in students' immediate experience (e.g., school or in the com-
munity) may be more likely to activate the goals, emotions, and personal
agency beliefs associated with caring behavior than would experiences
that focus on large, abstract, or far-away problems that are likely to be
seen as irrelevant, unsolvable, and /or emotionally uncompelling. In addi-
tion, experiences that explicitly attempt to link caring behavior with the
diversity of goals that may be achieved through such behavior (e.g., the in-
trinsic pleasure of helping others; positive feelings about one's self; a
sense of connectedness with other people; a sense of meaningful task ac-
complishment) are likely to produce more initial and enduring interest and
more meaningful engagement in prosocial activities than are programs
that fail to make these motivational connections. Similarly, interventions
that directly address the competing priorities that may inhibit caring be-
havior (e.g., a desire to avoid social obligations and commitments; a de-

sire to maintain resources for one's self) may be more successful in facili-
tating caring activity than those that ignore these potentially inhibiting
factors. Finally, educators who are not afraid to involve students in experi-
ences that elicit deep and genuine emotions reflecting concern for others
will probably be more successful in building and strengthening caring be-
havior patterns than will those who take a more "academic" (i.e., informa-
tion-giving) approach, or that rely on relatively superficial activities that
fail to produce meaningful emotional involvement.
The wide range of individual differences observed in the psychological
processes associated with caring competence suggests that efforts to pro-
mote caring behavior are likely to have more impact on some people than
others. In particular, it may be difficult to "motivate" some individuals at
the low end of the caring continuum to think and act more frequently on
behalf of others due to major deficits in one or more of the core psycho-
logical elements associated with caring behavior (e.g., weak social re-
sponsibility and resource provision goals; low levels of guilt and empathic
concern). The task of facilitating caring behavior in such individuals may
be even more formidable to the extent that the essential elements of the
caring "personality" are developed early in life (e.g., in the context of ear-
ly socialization experiences with caregivers and peers, perhaps with facil-
itating and constraining influences from inborn arousal patterns)
(Kochanska, 1994; Maccoby, 1992). It may be that, in some cases, all one
can realistically hope to do is to construct consequences for antisocial be-
havior that are sufficiently compelling to provide an "extrinsic" basis for
behaving in a socially cooperative manner.
Nevertheless, the extensive within-person variability observed across
measures and situations in the studies described in this chapter suggests
that most people have the capacity for caring behavior and will manifest
such behavior under certain circumstances. For example, almost everyone
- even many of those in the low caring group - indicated that they would
anticipate feeling some degree of concern for others, and might be willing
to act on behalf of others in at least some situations. Thus, although the
task may be more difficult in some cases than others, interventionists
should proceed with an optimistic sense of the possibilities for promoting
caring behavior in youth, recognizing that uncaring behavior usually re-
flects the dominance of more immediate self-focused priorities rather than
an inability to behave in a caring manner.
The research literature also suggests that interventions designed to pre-
vent socially irresponsible behavior (e.g., delinquent behavior; drinking
and driving; unprotected sexual activity; poor citizenship) may need to

concentrate first and foremost on facilitating an obligation motivational
pattern. Caring goals and emotions may also be appropriate targets of in-
tervention, but they appear to be more relevant for promoting prosocial
behavior than for inhibiting antisocial behavior. On the other hand, given
the motivational strength resulting from the simultaneous operation of
multiple goals and emotions, perhaps the most effective strategy for pro-
moting both caring competence and social responsibility in youth would
be to try to facilitate an optimal balance of caring and obligation motives
tailored both to the goals of the intervention and the personalities of the
individuals participating in the intervention.
References
Bandura, A. (1991). Social cognitive theory of moral thought and action. In W. M.
Kurtines & J. L. Gewitz (Eds.), Handbook of moral behavior and development
Vol.l: Theory (pp. 45-103). Hillsdale, NJ: Erlbaum.
Batson, C. D. (1990). How social an animal? The human capacity for caring.
American Psychologist, 45, 336—346.
Batson, C. D., Dyck, J. L., Batson, J. G., Powell, A. L., McMaster, M. R., & Grif-
fitt, C. (1988). Five studies testing two new egoistic alternatives to the empa-
thy-altruism hypothesis. Journal of Personality and Social Psychology, 55,
52-77.
Bergin, C. A. C. (1987). Prosocial development in toddlers: The patterning of
mother-infant interactions. In M. E. Ford & D. H. Ford (Eds.), Humans as self-
constructing living systems: Putting the framework to work (pp. 121-143).
Hillsdale, NJ: Erlbaum.
Cantor, N., & Harlow, R. (1994). Social intelligence and personality: Flexible life-
task pursuit. In R. J. Sternberg & P. Ruzgis (Eds.), Personality and intelligence
(pp. 137-168). New York: Cambridge University Press.
Chapman, M., Zahn-Waxier, C, Cooperman, G. Y., Iannotti, R. (1987). Empathy
and responsibility in the motivation of children's helping. Developmental Psy-
chology 23, 140-145.
Connell, J. P., & Wellbourn, J. G. (1991). Competence, autonomy, and relatedness:
Sroufe (Eds.) Self processes and development: The Minnesota symposium on
child psychology (vol. 23, pp. 43-78). Hillsdale, NJ: Erlbaum.
Damon, W. (1995). Greater expectations. New York: Free Press.
Eisenberg, N., & Miller, P. (1987). Empathy and prosocial behavior. Psychologi-
cal Bulletin, 707,91-119.
Eisenberg, N., & Mussen, P. H. (1989). The roots of prosocial behavior in chil-
dren. New York: Cambridge University Press.
Emmons, R. A. (1989). The personal striving approach to personality. In L. A.

Pervin (Ed.), Goal concepts in personality and social psychology (87—126).
Feldman, S., & Elliott, G. (Eds.) (1990) At the threshold: The developing adoles-
cent. Cambridge, MA: Harvard University Press.
Ford, D. H. (1987). Humans as self-constructing living systems: A developmental
perspective on behavior and personality. Hillsdale, NJ: Erlbaum.
Ford, M. E. (1992). Motivating human: Goals, emotions, and personal agency be-
liefs. Newbury Park, CA: Sage.
Ford, M. E., Chase, C , Love, R., Pollina, S., & Ito, S. (1991). Qualities associated
with caring behavior in adolescence: Goals, emotions, and personal agency be-
liefs. Program report submitted to the Lilly Endowment, Indianapolis, IN.
Ford M. E., & Ford, D. H. (Eds.) (1987). Humans as self-constructing living sys-
tems: Putting the framework to work. Hillsdale, NJ: Erlbaum.
Ford M. E., & Nichols, C. W. (Eds.) (1987). A taxonomy of human goals and
some possible applications. In M. E. Ford & D. H. Ford (Eds.), Humans as self-
constructing living systems: Putting the framework to work (pp. 289—311).
Ford, M. E., & Nichols, C. W. (1991). Using goal assessments to identify motiva-
tional patterns and facilitate behavioral regulation and achievement. In M. L.
Maehr & P. Pintrich (Eds.), Advances in motivation and achievement, Vol. 7:
Goals and self-regulatory processes (pp. 51-84). Greenwich, CT: JAI Press.
Ford, M. E., Wentzel, K. R., Siesfeld, G. A., Wood, D., & Feldman, L. (1986).
Adolescent decision making in real-life situations involving socially responsi-
ble and irresponsible choices. Paper presented at the annual meeting of the
American Educational Research Association, Los Angeles.
Ford, M. E., Wentzel, K. R., Wood, D., Stevens, E., & Siesfeld, G. A. (1989).
Processes associated with integrative social competence: Emotional and con-
textual influences on adolescent social responsibility. Journal of Adolescent Re-
search, 4, 405-425.
Frijda, N. H. (1988). The laws of emotion. American Psychologist, 43, 349-358.
Hoffman, M. L. (1982). Development of prosocial motivation: Empathy and guilt.
In N. Eisenberg (Ed.), The development of prosocial behavior (pp. 281-313).
New York: Academic Press.
Kochanska, G. (1994). Beyond cognition: Expanding the search for the early roots
of internalization and conscience. Developmental Psychology, 30, 20-22.
Krebs, D. L. (1975). Empathy and altruism. Journal of Personality and Social
Psychology, 32, 1134-1146.
Krumboltz, J. D., Ford, M. E., Nichols, C. W, & Wentzel, K. R. (1987). The goals
of education. In R. C. Calfee (Ed.), The study of Stanford the schools: Views
from the inside, Part II. Stanford University: Stanford, CA.
Maccoby, E. E. (1992). The role of parents in the socialization of children: An his-
torical overview. Developmental Psychology, 22, 1006-1017.
Markus, H., & Ruvolo, A. (1989). Possible selves: Personalized representations of

goals. In L. A. Pervin (Ed.), Goal concepts in personality and social psycholo-
gy (pp. 211-241). Hillsdale, NJ: Erlbaum.
Meece, J. L., Wigfield, A., & Eccles, J. S. (1990). Predictors of math anxiety and
its influence on young adolescents' course enrollment intentions and perfor-
mance in mathematics. Journal of Educational Psychology 82, 60-70.
Noddings, N. (1984). Caring: A feminine approach to ethics and moral education.
Berkeley, CA: University of California Press.
Noddings, N. (1992). The challenge to care in schools: An alternative approach to
education. New York: Teachers College Press.
Radke-Yarrow, M., Zahn-Waxier, C, & Chapman, M. (1983). Children's proso-
cial dispositions and behavior. In E. M. Hetherington (Ed.), Handbook of child
psychology Vol. 4: Socialization, personality, and social development (pp.
469-545). New York: Wiley.
Slavin, R. E. (1981). When does cooperative learning increase student achieve-
ment? Psychology Bulletin, 94, 429^45.
Slavin R. E. (1987). Developmental and motivational perspectives on cooperative
learning: A reconciliation. Child Development, 58, 1161-1167.
Wentzel, K. R. (1989). Adolescent classroom goals, standards for performance,
and academic achievement: An interactional perspective. Journal ofEducation-
al Psychology 81, 131-142.
Wentzel, K. R. (1991a). Social competence at school: Relation between social re-
1-24.
Wentzel, K. R.. (1991b). Relations between social competence and academic
Wentzel, K. R. (1993b). Does being good make the grade? Relations between aca-
demic and social competence in early adolescence. Journal ofEducational Psy-
chology, 85, 357-364.
Wentzel, K. R. (1993b). Social and academic goals at school: Motivation and
achievement in early adolescence. Journal of Early Adolescence, 13,4-20.
Wentzel, K. R. (1994). Relations of social goal pursuit to social acceptance, class-
room behavior, and perceived social support. Journal of Educational Psycholo-
gy, 86, 173-182.
7 Modeling and self-efficacy influences on
children's development of self-regulation
Dale H. Schunk and Barry J. Zimmerman
An essential feature of children's adjustment to school is their develop-

ment of academically-related self-regulation and motivation. The question
of how children acquire and internalize adult levels of self-regulatory
competence and motivation has fascinated scholars since the ancient
Greeks. Although philosophers, sociologists, and anthropologists have
long emphasized the importance of socialization influences, until recently
there has been relatively little detailed information about how such influ-
ences affect children's self-regulatory development. Since the late 1950s,
researchers in the social learning tradition (Bandura & Walters, 1963)
have hypothesized that children's exposure to socializing agents, especial-
ly adult or peer models, influences their behavioral and cognitive develop-
ment (e.g., formation of concepts, attitudes, preferences, standards for
self-reward and self-punishment). They found extensive evidence that
children readily induce and transfer concepts that underlie modeling se-
quences (Rosenthal & Zimmerman, 1978; Zimmerman & Rosenthal,
1974).
In recent years, social cognitive theorists shifted their attention to ado-
lescents' internalization of self-regulatory competence and studied how
youngsters learn to function independently from socializing agents in an
adaptive, generative, and creative manner. Bandura (1986) emphasized
the importance of a number of specific self-regulatory processes. Self-reg-
ulation refers to processes students use to activate and sustain cognitions,
behaviors, and affects, which are oriented toward the attainment of goals
(Zimmerman, 1989, 1990). Academic self-regulatory processes include
planning and managing time; attending to and concentrating on instruc-
tion; organizing, rehearsing, and coding information strategically; estab-
lishing a productive work environment; and using social resources effec-
tively. In addition, self-regulation incorporates such motivational
processes as setting performance goals and outcomes; holding positive
154
Influences on self-regulation 155
beliefs about one's capabilities; valuing learning and its anticipated out-
comes; and experiencing pride and satisfaction with one's efforts (Schunk
& Zimmerman, 1994).
This chapter focuses on the social and self origins of students' develop-
ment of self-regulatory skill with special emphasis on modeling and self-
efficacy beliefs. Modeling occurs when observers pattern their behaviors,
strategies, thoughts, beliefs, and affects after those of one or more models
(Schunk, 1987). Self-efficacy refers to beliefs about one's capabilities to
learn or perform behaviors at designated levels (Bandura, 1986, in press).
Recent research has demonstrated the effectiveness of modeling self-regu-
latory skills on students' academic achievement and associated self-effi-
cacy beliefs (Schunk & Zimmerman, 1994).
We initially present a theoretical overview of students' development of
self-regulatory competence followed by a social cognitive analysis of ma-
ture self-regulation based on Bandura's (1986) account. We then summa-
rize research on the roles of modeling and self-efficacy in the develop-
ment of self-regulatory competence. We conclude with some implications
of the theory and research for educational practice.
Theoretical background
Development of self-regulatory competence
According to a social cognitive perspective, students' academic compe-
tence develops initially from social sources of academic skill and subse-
quently shifts to self sources in a series of levels of skill as depicted in
Table 7.1 (Zimmerman & Bonner, in press). At the outset, novice learners
acquire learning strategies most rapidly from social modeling, tuition,
task structuring, and encouragement (Zimmerman & Rosenthal, 1974).
Table 7.1 A social cognitive analysis of primary influences on students' self-

regulatory development
Levels of development Social influences Self influences
Observational Modeling, verbal description
Imitative Social guidance and feedback
Self-controlled Internal standards,
self-reinforcement
Self-regulated Self-regulatory processes,
self-efficacy beliefs
Although many learners can induce the major features of learning strate-
gies from watching a model {observational level of academic skill), most
of them will require motoric performance experiences in order for them to
fully incorporate the skill into their behavioral repertoire. If the model
adopts a teaching role and provides guidance, feedback, and social rein-
forcement during imitative practice, the observer's motoric accuracy can
be improved further. During participant or mastery modeling (Bandura,
1986), the model repeats selected aspects of the strategy and guides enact-
ment based on the learners' imitative accuracy. An imitative level of acad-
emic skill is attained when the learner's performance approximates the
general form of the model. This does not mean that the observer is exactly
copying the model (an effect termed "response mimicry"). More often,
the learner emulates only the general pattern or style of the model, such as
the type of question a model asks, instead of duplicating the model's
words (Zimmerman & Rosenthal, 1974).
Although the source of learning of a skill is primarily social for the
first two levels of academic competence, it shifts to self-influences at
more advanced levels. The most apparent manifestation of a third or self-
controlled level of academic skill is a learner's ability to use the strategy
independently when performing transfer tasks. Students' use of a learning
strategy becomes internalized during this phase, but it remains dependent
on representational standards of a model's performance (i.e., covert im-
ages and verbal meanings) and the self-reinforcement that stems from
matching them (Bandura & Jeffery, 1973).
Socialization agents expect an even higher level of self-functioning
when students reach adolescence. A fourth of self-regulated level of acad-
emic skill is needed so learners can systematically adapt their learning
strategies to changing personal and contextual conditions (Bandura,
1986). At this level of academic competence, the learner can initiate use
of strategies, can incorporate adjustments based on contextual features of
the applied situation, and is motivated primarily by self-efficacy percep-
tions of enactive success. The learner chooses when to use a strategy and
varies its features self-regulatively, with little or no residual dependence
on the model.
In summary, a four level analysis of self-regulatory development ex-
tends from acquiring knowledge of learning skills (observation), to using
them (imitation), to their intemalization (self-control), and finally to their
adaptive use (self-regulation). Although this social cognitive model was
initially derived from research on children's socialization processes, it has
proven useful in guiding instructional efforts to teach students how to ac-

quire and eventually self-regulate their academic learning.
Social cognitive theory of self-regulation

Triadic reciprocality. Bandura's (1986) social cognitive theory views
human functioning in terms of interactions between behaviors, environ-
mental variables, and cognitions and other personal factors. Each of these
three classes of variables can influence the other and is in turn influenced
by it. For example, with respect to the link between cognitions and behav-
iors, much research shows that students' self-efficacy beliefs influence
achievement behaviors (choice of tasks, effort, persistence, achievement)
(Schunk, 1989). Conversely, students' behaviors also can alter efficacy be-
liefs. As students work on tasks they note their progress toward their
learning goals. Goal progress and accomplishment convey to students that
they are capable of performing well, which enhances self-efficacy for
continued learning.
Students' behaviors and classroom environments are similarly recipro-
cally dependent. In a typical instructional sequence, a teacher may present
information and direct students' attention to material on the board. Envi-
ronmental influence on behavior occurs when students respond to the
board without much conscious deliberation. Conversely, students' behav-
iors alter their environment. When students answer incorrectly, their
teacher may reteach the lesson in a different fashion rather than continue
with the original material.
Finally, cognitions and environments exert reciprocal effects. When
presented with the same difficult task, students with high self-efficacy
may view the task as a challenge and work diligently to master it, whereas
those with low efficacy may attempt to avoid it (cognition affects environ-
ment). The influence of environment on cognition is seen when teachers
provide feedback to students (e.g., "That's correct. You're getting good at
this.") that raises their self-efficacy and sustains their motivation to con-
tinue to improve their skills.
Self-regulatory processes. Social cognitive research has identified three

major classes of self-regulation: self-observation, self-judgment, and self-
reaction (Bandura, 1986; Kanfer & Gaelick, 1986). Self observation in-
volves deliberate attention to specific aspects of one's behavior (Bandura,
1986). Behaviors can be assessed on such dimensions as quantity, quality,
rate, and originality. When self-observation reveals goal progress, it can

motivate one to improve (Schunk, 1994). Often students with academic
problems are surprised to learn that they waste much valuable study time
on nonacademic distractions. Such knowledge can motivate students to re-
form their ways.
Self-observation is assisted with self-recording, where instances of
behavior are recorded along with their time, place, and frequency of oc-
currence (Mace, Belfiore, & Shea, 1989). In the absence of recording, ob-
servations may be forgotten or misinterpreted. The validity of self-obser-
vation is improved if behavior is observed at regular intervals and close in
time to its occurrence rather than sporadically and long after it occurs.
Self-observation is closely linked to a second self-regulatory process,
self-judgment, which refers to comparing present performance with a
standard. Self-judgments are affected by type and importance of standards
employed.
Standards may be stated in absolute or normative terms. The standard
for an absolute goal is fixed (e.g., write five pages in one hour). Norma-
tive standards are based on performances of others and often are acquired
by observing models (Bandura, 1986). Social comparison of one's perfor-
mances with those of others helps one evaluate the appropriateness of be-
havior. Social comparison becomes more likely when absolute standards
are not in effect or are unclear (Schunk, 1994).
Self-judgments are also affected by the importance of standards. Peo-
ple make progress judgments for behaviors they value. They may not as-
sess their performance or expend effort to improve their skills in areas
where they care little how they perform.
Standards are informative and motivational. Comparing one's perfor-
mance against standards informs one of progress. Writers who must com-
plete a 30 page chapter in a week know they are ahead of schedule if they
complete three pages during the first hour. The awareness that one is mak-
ing extraordinary progress enhances self-efficacy and sustains motivation
(Schunk, 1994).
Self-reaction refers to evaluations made of one's performance:
good/bad, acceptable/not acceptable, beyond/below expectation. Evalua-
tive reactions involve students' beliefs about their progress. The belief that
one is making acceptable goal progress, along with the anticipated satis-
faction of goal attainment, enhances self-efficacy, and sustains motiva-
tion. Negative evaluations will not decrease motivation if students believe
they are capable of improving (i.e., by working harder or using more ef-
fective strategies) (Schunk, 1994). Motivation is not enhanced if students
think they lack the ability to succeed and that more effort or better strate-
gy use will not help.
Self-reactions can be augmented by tangible rewards. Students who be-
lieve they are making good progress in studying for an exam may reward
themselves with two hours off from studying to watch a TV program. Tan-
gible rewards validate the perception of progress and raise self-efficacy
when they are linked to actual accomplishments.
These self-regulatory processes interact with one another. As students
observe aspects of their behavior, they judge them against goal standards
and react to those judgments. Their evaluations and reactions set the stage
for additional observations of the same behaviors or of others. These
processes also interact with environmental factors (Zimmerman, 1989).
Students who judge their task progress as inadequate may react by re-
questing teacher assistance. Teachers may teach students a better strategy,
which students then use to promote learning. This dynamic interaction of
self-regulation processes is one of its central features (Schunk & Zimmer-
man, 1994).
Modeling processes. Modeling processes are important socialization

components of self-regulation. Modeling can serve different functions: in-
hibition/disinhibition, response facilitation, observational learning. Inhi-
bition/disinhibition means that observing a model can strengthen or weak-
en behavioral inhibitions. Models who perform threatening or prohibited
activities without experiencing negative consequences may influence ob-
servers to perform the behaviors themselves; models who are punished
may inhibit observers' responding. Inhibitory and disinhibitory effects
stem from the information conveyed to observers that similar conse-
quences are likely should they act accordingly.
Response facilitation denotes modeled actions serving as social
prompts for observer behavior. Response facilitation occurs when we fol-
low the actions of the crowd. A student who walks into a room where oth-
er students are studying quietly is apt to respond to this prompt and be-
have the same way.
Both inhibitory/disinhibitory and response facilitation effects are im-
portant for self-regulation because students may regulate their activities
based on the social cues they observe and the consequences of modeled
actions. More important for education, however, are observational learn-
ing effects. Observational learning through modeling occurs when ob-
servers display new behaviors that, before modeling, had no probability of
occurrence, even when motivational inducements are offered (Bandura,
1986; Schunk, 1987). To learn observationally, students must attend to a

model, code the information for retention, be capable of producing the
demonstrated response pattern, and be motivated to perform imitatively
(Bandura, 1986). An important form of observational learning occurs
through cognitive modeling, which incorporates modeled explanations
and demonstrations with verbalizations of the model's thoughts and rea-
sons for performing the actions (Meichenbaum, 1977).
The functional value of behavior - whether modeled behaviors result
in success or failure, reward or punishment - exerts strong effects on ob-
server modeling. Modeled behaviors are likely to be performed if they
previously led to rewarding outcomes, but are unlikely if they resulted in
punishment.
Modeling experiences fulfill informational and motivational functions.
Vicarious consequences indicate the motivational value of behavior to ob-
servers (Bandura, 1986); antecedent actions inform observers about what
should be done to attain them. Most social situations are structured so that
the appropriateness of behaviors depends on such factors as age, gender,
or status. By observing modeled behaviors and their consequences, people
formulate expectations about the likely outcomes of actions. Vicarious
consequences create outcome expectations about which behaviors will be
rewarded and which punished.
Perceived similarity between model and observer is hypothesized to be
an important source of information to determine behavioral appropriate-
ness and formulate outcome expectations (Schunk, 1987). The more alike
observers are to models, the greater is the probability that similar actions
by observers are socially appropriate and will produce comparable results.
Model attributes often are predictive of the functional value of behavior.
Similarity should be especially influential in situations where observers
have little information about functional value. Modeled behaviors on tasks
that observers are unfamiliar with or those that are not immediately fol-
lowed by consequences may be highly susceptible to influence by attribute
similarity.
Vicarious consequences also motivate observers. These effects depend
in part on self-efficacy. Similarity to models constitutes an important
source of vicarious information for gauging one's efficacy. Observing
similar others succeed can raise observers' efficacy and motivate them to
try the task based on the assumption that if others can succeed, they can as
well. Observing similar others experiencing difficulty may lead observers
to doubt their own capabilities and undermine their motivation to try the
task. Model attributes provide information about what one can do. Simi-
larity is highly influential in situations where individuals have experi-

enced difficulties and hold doubts about performing well.
Self-Efficacy
Sources, effects, and consequences. Self-efficacy is hypothesized to in-
fluence choice of tasks, effort expenditure, persistence, and achievement
(Bandura, 1986, in press; Schunk, 1989; Zimmerman, 1995) (See also
Erdley, this volume, for ways in which self-efficacy is related to peer-di-
rected aggression). Compared with students who doubt their learning ca-
pabilities, those holding a sense of efficacy for acquiring a skill or per-
forming a task participate more readily, work harder, persist longer when
they encounter difficulties, and achieve at a higher level.
Learners obtain information to appraise their self-efficacy from their
performance accomplishments, vicarious (observational) experiences,
forms of persuasion, and physiological reactions. Students' own perfor-
mances offer reliable guides for assessing self-efficacy. Successes raise
efficacy and failures lower it, although once a strong sense of efficacy de-
velops, an occasional difficulty typically has little impact (Schunk, 1989).
Students socially acquire efficacy information by comparing their per-
formances with those of others. Similar others offer the best basis for
comparison (Schunk, 1987). Students who observe similar peers perform
a task are apt to believe that they, too, are capable of accomplishing it. In-
formation acquired vicariously typically has a weaker effect on self-effic-
acy than performance-based information because the former can be out-
weighed by subsequent failures.
Learners often receive from teachers, parents, coaches, and peers, per-
suasive information that they are capable of performing a task ("You can
do this"). Positive persuasive feedback enhances self-efficacy, but this in-
crease will be temporary if subsequent efforts turn out poorly. Students
also acquire efficacy information from physiological reactions (e.g.,
sweating, heart rate). Symptoms signaling anxiety may convey that one
lacks skills; conversely, students who experience less agitation feel more
efficacious about performing well.
Information derived from these sources does not influence self-effica-
cy automatically but rather is cognitively appraised (Bandura, 1986, in
press). Learners weigh and combine the contributions of many factors in-
cluding perceptions of ability, task difficulty, amount of effort expended,
amount and type of assistance from others, perceived similarity to models,
and persuader credibility (Schunk, 1989).
Self-efficacy is important, but not the only influence on achievement

behavior. High self-efficacy will not produce competent performances
when requisite knowledge and skills are lacking. Outcome expectations
are influential because students engage in activities they believe will lead
to positive outcomes. Similarly, perceived value (importance attached to
learning or what use will be made of what one learns) affects behavior be-
cause learners show little interest in activities they do not value. When
students possess adequate skills, expect positive outcomes, and value
what they are learning, self-efficacy is hypothesized to influence the
choice and direction of much achievement behavior.
Self-efficacy and self-regulation. Effective self-regulation depends on a

sense of self-efficacy for using skills to achieve mastery (Bandura, 1986,
in press; Bouffard-Bouchard, Parent, & Larivee, 1991; Schunk, 1994;
Zimmerman, 1989). As students work on a task, they compare their per-
formances to their goals. Self-evaluations of progress enhance self-effica-
cy and keep students motivated to improve. Students who feel efficacious
about learning or performing well are apt to implement various effective
self-regulatory strategies, such as concentrating on the task, using proper
procedures, managing time effectively, seeking assistance as necessary,
and monitoring performance and adjusting strategies as needed (Schunk,
1994; Zimmerman, 1994).
Although low self-efficacy is detrimental, effective self-regulation
does not require that self-efficacy be extremely high. Salomon (1984)
found that lower self-efficacy led to greater mental effort and better learn-
ing than when efficacy was higher. Assuming that learners feel efficacious
about surmounting problems (a very low sense of efficacy is not motivat-
ing), holding some doubt about whether one will succeed may mobilize
effort and effective use of strategies more than will feeling overly confi-
dent.
Research Evidence
In this section we present a limited review of research on the social and
self origins of self-regulatory competence. These categories are distin-
guished according to the extent they require influences in the social envi-
ronment for development. Social origins include observational learning
and imitative learning; self origins comprise self-instruction in achieving
self control and self-regulated learning through goal setting, self-monitor-
ing, self-evaluation, help seeking, and time management. This review is

selective; readers who want a more comprehensive treatment should con-
sult other sources (Bandura, 1986, in press; Schunk & Zimmerman, 1994;
Zimmerman & Schunk, 1989).
Social origins of self-regulatory competence

Much of what we know about social origins of self-regulatory competence
derives from theory and research on modeling. The research in this sec-
tion on observational and imitative learning through modeling focuses on
how students develop skills and access information about their competen-
cies from exposure to adult and peer models. There is a vast literature dis-
cussed in other sources on how models affect personal relationships be-
tween students and significant adults and peers and the implications of
these relationships for school adjustment (Bandura, 1986; Rosenthal &
Bandura, 1978). The role of modeling also figures prominently in the cur-
rent educational emphasis on mentoring and apprenticeships (e.g., Bailey,
1993).
Observational learning. Models are important sources for the initial de-
velopment of self-regulation: an observational level of skill. By observing
models, students acquire knowledge and strategies that they subsequently
apply as they work on tasks. Modeled displays also convey to observers
that they can succeed if they follow the same sequence of actions. The be-
lief that one knows what to do to perform a task raises self-efficacy, which
is increased further as observers work on the task and experience success
(Schunk, 1989).
An important means of developing an observational level of compe-
tence is through cognitive modeling. Schunk (1981) gave children who
had encountered difficulty in mathematics either cognitive modeling or
didactic instruction. In the modeling treatment, children observed an adult
model verbalize division operations while applying them to problems. The
didactic treatment consisted of children reviewing instructional pages that
portrayed the solution of division problems step-by-step. Both cognitive
modeling and didactic instruction led to significant increases in self-effi-
cacy, skill, and persistence, but modeling resulted in significantly higher
division skill performance. Results of a path analysis showed that self-ef-
ficacy had an direct effect on both persistence and skill.
Perceived similarity to models in important attributes can raise ob-
servers' self-efficacy and motivate them to try the task because they are
likely to believe that if others can succeed, they can as well (Schunk,
1987). One way to vary similarity is through the use of coping and mas-
tery models. Coping models initially demonstrate the typical behavioral
deficiencies and possibly fears of observers, but gradually improve their
performance and gain self-confidence. These models illustrate how effort
and positive thoughts can overcome difficulties. Mastery models demon-
strate faultless performance from the outset (Schunk, 1987).
Schunk and Hanson (1985) compared peer mastery and coping models
with adult teacher models and no models. Peer mastery models solved
subtraction problems correctly and verbalized statements reflecting high
self-efficacy and ability, low task difficulty, and positive attitudes. Peer
coping models initially made errors and verbalized negative statements,
but then verbalized coping statements (e.g., "I need to pay attention to
what I'm doing") and eventually verbalized and performed as well as mas-
tery models. Peer models increased self-efficacy and skill better than the
teacher model or no model; teacher-model children outperformed no-
model students. Although teacher models can teach students self-regulato-
ry skills, students' self-efficacy beliefs for learning may be aided better by
observation of similar peers. In turn, self-efficacy can raise motivation for
skill improvement (Schunk, 1989).
Schunk, Hanson, and Cox (1987) found that observing peer coping
models enhances children's self-efficacy and skillful performance more
than does observing peer mastery models. Unlike the Schunk and Hanson
(1985) study, these authors used a task (fractions) with which children had
no prior successful performances. Coping models may be more beneficial
when students have little task familiarity or have encountered previous
learning difficulties. Schunk et al. (1987) also showed that multiple mod-
els (coping or mastery) promote outcomes as well as a single coping mod-
el and better than a single mastery model.
Models can convey abstract rules and concepts for self-regulation,
such as self-evaluative standards. Bandura and Kupers (1964) exposed
children to a model demonstrating stringent or lenient standards while
playing a bowling game. Children exposed to high-standard models were
more likely to reward themselves for high scores and less likely to reward
themselves for lower scores compared with subjects assigned to the low-
standard condition. Davidson and Smith (1982) had children observe a su-
perior adult, equal peer, or inferior younger child set stringent or lenient
standards while performing a pursuit rotor task. Children who observed a
lenient model rewarded themselves for lower scores than those who ob-
served a stringent model. Children's self-reward standards were lower than
those of the adult, equal to those of the peer, and higher than those of the
younger children. Age similarity may have led children to believe that
peer standards were appropriate for them.
Imitative learning. Models can provide social evaluative cues, feedback,

and assistance to help observers achieve an imitation level of motoric
competence as well. France-Kaatrude and Smith (1985) had first and
fourth graders perform a pursuit-rotor task, and children could compare
their performances with a peer of higher, equal, or lower competence.
Compared with children offered social comparisons with superior or infe-
rior peers, those allowed to compare with a similarly-performing peer
compared more often, demonstrated greater task persistence, and took
fewer self-rewards.
Self-modeling. Self-modeling, which involves watching one's own per-

formances, is another effective method of developing imitative compe-
tence (Dowrick, 1983). Typically, one is videotaped while performing a
task and subsequently views the tape. Tapes allow for review and are espe-
cially informative for tasks one cannot watch while performing. When
performances contain errors, commentary from a knowledgeable individ-
ual during tape review helps to prevent performers from becoming dis-
couraged. The expert can explain how to execute the behavior better the
next time. Tapes can convey to observers that they are becoming more
skillful and can continue to make progress, which raises self-efficacy. In
support of these points, Schunk and Hanson (1989) videotaped children
solving problems and showed them their tapes. Subsequent self-modeling
benefits were obtained as these children displayed higher self-efficacy
and motivation than children who had been taped but did not observe their
tapes, and those who had not been taped.
To fully achieve an imitative level of competence, adults must fade so-
cial and instructional supports and encourage students to work on tasks
on their own. This should be done gradually, as students abstract the un-
derlying learning strategy and receive progress feedback.
Schunk and Swartz (1993) explored the effects of learning goals and
progress feedback on children's self-efficacy, self-regulatory use of writ-
ing strategies, and writing skill. Students received writing instruction and
practice on different types of paragraphs over several sessions. They were

taught a five-step writing strategy (e.g., choose a topic to write about, pick
the main idea). Once children learned the strategy, they practiced applying
it to paragraphs independently as adult support was faded out.
There were four experimental conditions: process (learning) goal,
process goal plus progress feedback, product (performance) goal, general
goal (instructional control). Process-goal and process-goal plus progress-
feedback children received instructions emphasizing the goal of learning
to use the strategy to write paragraphs. Product-goal students were told
their goal was to write paragraphs; general-goal students were advised to
do their best. Children assigned to the process-goal plus progress-feed-
back condition periodically received verbal feedback from the adult
teacher that linked their use of the strategy with improved writing perfor-
mance (e.g., "You're doing well because you followed the steps in order").
Feedback was given contingent on students using the strategy properly.
Schunk and Swartz felt that this feedback would raise self-efficacy and
motivation by informing students of their learning progress.
The results showed that the process-goal plus feedback condition was
the most effective and that there also were some benefits of providing a
process goal alone. Process-goal plus feedback students generally outper-
formed product- and general-goal students on self-efficacy and writing
skill. The former subjects also reported and demonstrated the highest
strategy use during the test sessions. Gains made by process-goal plus
feedback students were maintained after six weeks and generalized to
types of paragraphs on which students had received no instruction.
Self origins of self-regulatory competence

In this section we consider further how self-control and self-regulatory
level skills develop in students. We discuss the roles of self-verbalization,
goal setting, self-monitoring, self-evaluation, help seeking, and time man-
agement. It is true that each of these can be affected by social and instruc-
tional factors, and in some of the studies we discuss subtle or more direct
social influences affecting students' self-regulatory activities. At the same
time, there is much research showing the prominent role that students play
in the development of their skills and perceptions of capabilities (Schunk,
1989), ranging from studies where students set their own goals (Schunk,
1985) to research exploring students' verbalizations as they work on
achievement tasks (Schunk & Gunn, 1986). In general, then, we can say
that research on self origins addresses factors that are less dependent on
the social environment for development than the factors discussed in the
preceding section.
Self-controlled learning. A process that can foster a self-controlled use

of strategies on one's own is verbalization of the strategy as it is applied.
Self-directed verbalization helps students attend to important task features
and disregard irrelevant information. Self-verbalization also promotes en-
coding, such as when students must detect and integrate information
needed to solve problems (Diefenderfer, Holzman, & Thompson, 1985).
As a form of rehearsal, verbalization helps to instate a strategy; and the
belief that one knows and can use a strategy to improve performance can
raise self-efficacy and enhance motivation for strategy use (Schunk,
1989).
Schunk & Rice (1984) presented language-deficient children in grades
two to four with listening-comprehension instruction. Half of the children
in each grade verbalized strategic steps before applying them to questions;
the other half applied but did not verbalize the steps. Strategy verbaliza-
tion led to higher self-efficacy across grades, and promoted performance
among all but the second graders. Perhaps the demands of verbalization
interfered with the strategy encoding and retention.
In a follow-up study (Schunk & Rice, 1985), children in grades four
and five with reading-comprehension difficulties received instruction and
practice. Within each grade, half of the subjects verbalized a strategy be-
fore applying it. Strategy verbalization led to higher reading comprehen-
sion, self-efficacy, and ability attributions for success across grades.
Verbalization may be helpful, but to become fully self-regulating it
must be internalized as private speech (Berk, 1986). In Meichenbaum's
(1977) self-instructional training, for example, verbalizations made by
participants while working on the task eventually are faded to whispers
and covert (silent) instructions. Schunk and Cox (1986) tested the effect
of faded verbalization on self-efficacy and achievement. Students with
learning disabilities verbalized aloud subtraction solution steps and their
application to problems. Some students verbalized continuously; others
verbalized during the first half of the training program but then were ad-
vised to continue using the strategy but not verbalize aloud (discontinued
verbalization); and students in a third condition did not verbalize. Contin-
uous verbalization led to higher self-efficacy and skill than did the other
two conditions. When instructed not to verbalize aloud, discontinued ver-
balization students might have had difficulty internalizing the strategy,

and not used covert instructions to regulate their performances.
Goal setting. Goal setting is hypothesized to be an important cognitive

process affecting motivation and self-regulation (Bandura, 1988; Locke &
Latham, 1990; Schunk, 1989; see also Berndt & Keefe; Ford; Kupersmidt
et al.; Wentzel, this volume, for other perspectives on goals). Students
who set a goal on their own or have one suggested to them (e.g., by teach-
ers) are apt to make a commitment to attempt to attain it - which is neces-
sary for goals to affect performance - and to experience a sense of effica-
cy for attaining it. As they work at the task, they engage in self-regulatory
activities they believe will help them attain it; for example, they use effec-
tive strategies, monitor their progress, manage their time, and seek assis-
tance when needed. Self-efficacy is substantiated as learners observe goal
progress, which conveys that they are becoming skillful (Elliott & Dweck,
1988). Providing students with feedback also raises self-efficacy (Ban-
dura & Cervone, 1983; Schunk & Swartz, 1993). Heightened self-efficacy
sustains motivation and leads to further development of self-regulatory
skills.
Motivational benefits of goals depend on their properties: proximity,
specificity, difficulty (Bandura, 1986, 1988; Locke & Latham, 1990).
Proximal (short term) goals result in greater motivation than distant (long-
term) goals because it is easier to gauge progress toward a short term goal.
Specific goals improve motivation by focusing the learner's effort, and
difficult but attainable goals enhance motivation more than easy goals.
Self-regulatory benefits of goal setting have been obtained in several stud-
ies. Bandura and Schunk (1981) presented children with seven sets of sub-
traction material. Some children pursued a proximal goal of completing
one set during each of seven instructional sessions; a second group was
given a distant goal of completing all sets by the end of the last (seventh)
session; a third group was advised to work productively (general goal).
Proximal goals heightened self-efficacy, skill, motivation, and self-regula-
tory strategy use during instruction. The distant goal resulted in no bene-
fits compared with the general goal. These findings support the idea that
when students can gauge their goal progress, the perception of improve-
ment enhances efficacy. Assessing progress toward a distant goal is more
difficult, and uncertainty about learning will not instill a sense of efficacy
for skill improvement.
Schunk (1983a) compared the effects of social comparative informa-
tion with those of goal setting during mathematical instructional sessions.
Half of the children were given performance goals each session; the other
half were advised to work productively. Within each goal condition, half
of the subjects were told the number of problems that other similar chil-
dren had completed (which matched the session goal) to convey that the
goals were attainable; the other half were not given this social comparison
information. Goals enhanced self-efficacy; social comparison information
promoted motivation during the sessions. Students who received both
goals and comparative information demonstrated the highest skill. These
results suggest that providing children with a goal and social comparison
information indicating that it is attainable increases self-efficacy for
learning, which contributes to more productive performance during in-
structional sessions and greater skill acquisition.
To test the effects of goal difficulty, Schunk (1983b) provided children
with either a difficult (but attainable) or easier goal of completing a given
number of problems each session. To enhance motivation, half of the sub-
jects in each goal condition were told that they could attain the goal (e.g.,
"You can work 25 problems"). The other half received social comparison
information indicating that similar children completed that many prob-
lems. Difficult goals enhanced motivation and division skill; direct goal-
attainment information promoted self-efficacy.
Allowing students to set goals may enhance goal commitment and self-
efficacy. Schunk (1985) provided subtraction instruction to sixth-grade
students with learning disabilities. Children who set their own perfor-
mance goals and those who had goals suggested to them demonstrated
greater motivation and self-regulated strategy use than students who
worked without goals; self-set goals led to the highest self-efficacy and
skill.
Goal setting also has proven effective over longer periods. Morgan
(1985) found that proximal goals raise academic performance and intrin-
sic interest among college students over an academic year. Although self-
efficacy was not assessed the results suggest that goal proximity may af-
fect self-regulatory processes associated with studying over time. Further
investigation is needed of students' goal setting and self-regulation of mo-
tivation over longer periods and of how self-efficacy changes as they
progress toward future goals.
Earlier we discussed the benefits of providing goal progress feedback.
Especially when it is difficult for students to determine whether they are
making progress toward goal attainment such explicit feedback can raise
self-efficacy and contribute to greater self-regulated use of learning
strategies and achievement. The research by Schunk and Swartz (1993)
supports the point that goal progress feedback can enhance maintenance
and generalization of self-regulated strategy use and writing achievement.
Self-monitoring. Self-monitoring is an integral component of the self-

regulation process (Bandura, 1986; Kanfer & Gaelick, 1986; Zimmerman
& Martinez-Pons, 1990). Monitoring allows one to assess progress toward
goals. The perception of progress builds self-eflficacy and motivates sus-
tained self-regulatory efforts. Monitoring often is assisted by self-record-
ing aspects of one's behaviors.
Self-monitoring of progress can enhance self-efficacy and promote
self-regulation. Sagotsky, Patterson, and Lepper (1978) had fifth- and
sixth-graders periodically monitor their performances during mathemati-
cal instruction and record whether they were working on appropriate ma-
terial. Other students set daily performance goals; those in a third condi-
tion received self-monitoring and goal setting. Self-monitoring increased
achievement and the amount of time students spent engaged with the task;
goal setting had minimal effects. For goal setting to affect performance, it
may first be necessary to teach students to set challenging but attainable
goals.
Schunk (1983c) provided instruction to children who had experienced
difficulties acquiring subtraction skills. Self-monitoring students re-
viewed their work at the end of each instructional session and recorded the
number of pages of problems they completed. External-monitoring sub-
jects had their work reviewed by an adult who recorded the number of
pages completed. No-monitoring subjects received the instruction and
practice, but did not record pages or have them recorded.
Self- and external monitoring promoted self-efficacy, skill, and self-
initiated persistence solving problems, better than did no monitoring. The
two monitoring conditions did not differ on any measure. Benefits of
monitoring did not depend on children's performances during the instruc-
tional sessions; the three conditions completed comparable amounts of
work. The results suggest that monitoring of progress, rather than the
agent, enhanced children's perceptions of their learning progress and self-
efficacy for improvement.
Self-evaluation. Self-evaluation comprises self-judgments of present

performance by comparing it to one's goal and self-reactions to those
judgments in terms of its acceptability, noteworthiness, and so forth. Posi-
tive self-evaluations lead students to feel efficacious about learning and
motivated to continue to work diligently because they believe they are ca-
pable of making further progress (Schunk, 1989). Low self-judgments of
progress and negative self-reactions will not necessarily diminish self-ef-
ficacy and motivation if students believe they can succeed but that their
present approach is ineffective (Bandura, 1986). They may alter their self-
regulatory processes by working harder, persistirtg longer, adopting what
they believe is a better strategy, or seeking help from teachers and peers.
These and other self-regulatory activities are likely to lead to success
(Zimmerman & Martinez-Pons, 1990).
An emphasis on self-evaluation during task engagement helps to devel-
op self-regulatory skills. Research with children during learning of mathe-
matical skills (Schunk & Hanson, 1985; Schunk et al., 1987) and writing
skills (Schunk & Swartz, 1993) shows that measures of self-efficacy for
learning or improving skills collected before receiving instruction predict
subsequent motivation and skill acquisition. Masters and Santrock (1976)
found that preschool children who verbalized self-judgmental statements
during an effortful handle-turning task (e.g., "I'm really good at this")
persisted longer than children who verbalized self-critical or neutral state-
ments.
Bandura and Cervone (1983) obtained benefits of goals and self-evalu-
ative feedback. College students pursued a goal of increasing motor-skill
performance by 40% over baseline; others were given feedback indicating
they increased performance by 24%; and those in a third condition re-
ceived goals and feedback. Goals plus evaluative feedback had the
strongest effect on performance and self-efficacy, which predicted subse-
quent effort. Bandura and Cervone (1986) gave subjects a goal of 50%
improvement and false feedback indicating they increased by 24%, 36%,
46%, or 54%. Self-efficacy was lowest for the 24% group and highest for
the 54% condition. Subjects then indicated new goals and performed the
task. Effort was positively related to self-set goals and self-efficacy across
conditions. A measure of self-evaluation (self-satisfaction with perfor-
mance) showed that the greater the dissatisfaction and the higher the self-
efficacy, the stronger was subsequent effort expenditure.
Schunk (in press) investigated how goals and self-evaluation affected
motivation, self-regulated strategy use, self-efficacy, and achievement.
Fourth-graders received instruction and practice on fractions and either a
goal of learning how to solve problems (learning goal) or a goal of solving
problems (performance goal). In the first study, half of the students in each
goal condition assessed their problem-solving capabilities. The learning
goal with or without self-evaluation or the performance goal with self-eval-
uation led to higher self-efficacy, skill, motivation for self-regulated strate-

gy use, and learning goal orientation, than did the performance goal with-
out self-evaluation. In the second study, all students evaluated their
progress in skill acquisition. The learning goal led to higher motivation,
self-regulated strategy use, self-efficacy, fractions skill, self-evaluations,
self-satisfaction with progress in learning, and learning goal orientation.
Help seeking. Another important origin of self-regulatory competence

comes through students' efforts to seek help when they are baffled or oth-
erwise uncertain about the appropriateness or accuracy of their work. As
Zimmerman (1989, 1990) notes, self-regulated learners are cognitively,
metacognitively, and motivationally active. When confronted with a diffi-
cult task that requires assistance, self-regulated learners seek help from
knowledgeable persons (Newman, 1994). Help-seeking bears a positive
relation to academic achievement (Newman & Schwager, 1992).
A model of adaptive help-seeking begins with the student's awareness
of a lack of understanding (Newman, 1994). The student then considers
available information in deciding about the necessity for, target of, and
content of the request. The help seeker expresses a need for help that is
appropriate under the prevailing conditions, and finally processes the as-
sistance in such a way as to improve subsequent performance.
Research has explored the relation of various individual characteristics
to help-seeking. The most consistent findings are in the area of academic
goals. Research shows that children characterized as high in intrinsic mo-
tivation for learning are more likely to seek help than are students with
lower intrinsic motivation (Newman, 1994). Two dimensions of intrinsic
motivation bear especially strong relations to help-seeking: striving for in-
dependent mastery and preference for challenge. Among all students, the
greater the preference for challenge, the more likely students are to seek
help. This relates to the importance of difficult but attainable goals for
motivation. Striving for independent mastery refers to the extent of depen-
dence on the teacher. For young children, the greater the dependence on
the teacher, the greater the likelihood of seeking help; by middle school,
however, this is reversed such that the stronger the striving for mastery, the
greater the likelihood of seeking help.
Although help-seeking is self generated, it depends on the social envi-
ronment and can be affected by social factors, especially the use of mod-
els. Teachers can model the process of seeking help by working on a task,
getting stuck, deciding to ask for assistance, and posing a question. Show-
ing children videotapes of other children engaged in a task and seeking as-
sistance also can minimize potential inhibitions against seeking help

(Newman, 1994). From a self-efficacy perspective, we might expect that
knowing one can seek help if one is unsure about what to do should raise
self-efficacy for performing well, but research is needed on this point.
Time management. The purposive planning, monitoring, and manage-

ment of study time is another important source of students' skill in self-
regulating their academic performance. To succeed in classes with implic-
it and explicit deadlines, students must plan their study time, feel
efficacious about managing their time, and monitor their progress.
Britton and Tesser (1991) investigated the impact of college students'
time management skills on cumulative grade-point-average (GPA). To
evaluate the possibility that time management is merely an aspect of intel-
ligence, they included the Scholastic Aptitude Test (SAT) as a traditional
measure of intelligence. They planned to compare the separate contribu-
tions of time management skills and intelligence to the GPA. A factor
analysis of the time management scale revealed three components: time
attitude, short- and long-range planning. The time attitude items revealed
how students felt about their efficiency of time use, their control of time
use, and their skill in time self-monitoring. Britton and Tesser noted that
these time attitude measures were similar to measures of self-efficacy. For
example, the time attitude component includes items that directly measure
the feeling of being in charge of one's time and being able to say "no" to
people, in order to concentrate on homework. Britton and Tesser found
that only the time attitude and short-range planning components of their
scale were significant predictors of cumulative GPA. Interestingly, the ef-
fects of short-range planning and time attitude were found to be indepen-
dent of and stronger predictors of GPA than the SAT. A sense of self-effi-
cacy for managing study time is essential for effective action.
Students' perceptions of efficacy in controlling study time can have af-
fective and somatic benefits as well as cognitive benefits. Macan, Sha-
hani, Dipboye, and Phillips (1990) developed a questionnaire to assess
time management behaviors and attitudes of college students. One of four
subscales in this instrument (Perceived Control of Time) involved judg-
ments of self-efficacy. This subscale was similar to Britton and Tesser's
(1991) time attitude items, and was the most predictive of student out-
comes. Students who perceived greater control of their time reported sig-
nificantly higher performance evaluations, greater work and life satisfac-
tions, less role ambiguity and overload, and fewer job-induced and
somatic tensions.
Students can be taught to improve the effectiveness of their self-moni-

toring (Mount & Terrill, 1977). Undergraduate volunteers were assigned
to various self-recording conditions in which there were two target behav-
iors: the amount of time that students studied for class and the amount of
time they experienced "guilt" about studying when they were not engaged
in it. The type of time measured did not influence the effectiveness of
these self-monitoring results differentially. Students who used note cards
and graph paper achieved significantly higher final examination grades
than those who used either recording form alone. The authors concluded
that the superiority of the combined method of self-monitoring was due to
its provision of more continuous feedback than either method alone.
These results show that instructions to self-monitor time use on recording
forms affects college students' achievement.
Together these studies indicate that students' self-regulation of acade-
mic study time through time planning, self-efficacy beliefs, and self-mon-
itoring significantly improves academic attainment. These findings sup-
port predictions of social cognitive theory and research results on
academic self-regulation (Bandura, in press; Schunk, 1989; Zimmerman,
1989).
Implications for practice

The ideas we have presented suggest many potential implications for edu-
cational practice. Teachers should make greater use of models in the class-
room by emphasizing the importance of cognitive modeling where models
verbalize their thought processes, in addition to the steps they perform, as
they work on a task (Zimmerman & Kleefeld, 1977). Coping models can
help relieve students' fears and build their confidence by verbalizing cop-
ing statements (e.g., "I have to pay better attention") and progress state-
ments ("I'm doing better").
Models are teachers or peers who explain and demonstrate skills and
strategies in front of classes, but there are other ways to use models in the
classroom. One way is in cooperative groups, in which a small number of
students work jointly on a task. Responsibilities are divided so each group
member is responsible for something. Groups typically are set up so that
each member must master the skills and the group is not allowed to pro-
ceed until such skill mastery occurs. The characteristics of effective
groups have been documented (Cohen, 1994). For our purposes it is es-
sential that students serve as models for one another. A good way to do
this is for each student to work on some aspect of the task, and then ex-
plain it to the other group members after he or she has mastered it. This
type of positive peer model teaches skills and raises others' self-efficacy.
Teachers can develop student's self-regulatory skills by teaching them
effective strategies, emphasizing effort and persistence, teaching students
to set process goals, providing feedback on goal progress, monitoring skill
acquisition, and periodically evaluating skill development. However, self-
regulative mastery requires more than observational learning experience,
it involves practicing those components. If teachers can incorporate many
of these procedures into regular classroom instructional exercises, stu-
dents will have opportunities to develop their self-regulatory skills. Teach-
ers who have students set goals and monitor their goal progress are apt to
foster these skills in students.
Students need to practice skills at home and in other contexts as well as
school environments. Bruner (1985) has noted that learning to play chess,
work mathematical problems, and play the flute involve such common
skills as attention, memory, and maintaining frustration tolerance. Stu-
dents should seek ways that self-regulatory competencies they acquire in
school can be used elsewhere. Thus, goal setting learned in the context of
spelling instruction may come in handy in track as students set goals for
lowering their times at particular distances.
We have not discussed the role of parents in this chapter because we
felt to do so would take us beyond our intended scope; however, parents
are important in the development of their children's self-regulatory com-
petencies. One way they can have a major impact is by systematically
modeling skills they want their children to display. When models act one
way and verbalize another type of behavior, children are more strongly in-
fluenced by the modeled behaviors than by the preaching (Bryan & Wal-
bek, 1970). If parents want children to set goals, then parents should set
goals themselves and assist children with goal setting. If parents want
children to develop effort and persistence, parents must display those
qualities and not give up readily on tasks.
Parents also can assist students in the selection of friends and others
with whom they share academic interests and can become study partners.
Friends strongly impact desirable conduct (Berndt & Keefe, 1992). Stu-
dents who associate with academically-oriented peers are likely to devel-
op more self-regulatory study skills associated with studying. By suggest-
ing to their children friends with whom they might study and helping to
arrange such opportunities, parents can help to develop children's self-
regulation. To the extent that they have the resources for making such a
choice, parents can aid their children's self-regulation through selection of
schools or teachers.
Conclusion
In this chapter, we have discussed the roles of modeling and self-efficacy
in the development of students' self-regulation to include social and self
origins of self-regulatory competence. Acquisition of self-regulatory
skills through modeling and social learning experiences occurs in a series
of levels. We believe that educators and parents can do much to help chil-
dren develop a self-regulatory level of academic skill through modeling,
socially assisted motoric practice, and use of self-processes (self-verbal-
ization, goal setting, self-monitoring, self-evaluation, help-seeking, and
time management), and have suggested some ways to do this. More re-
search is needed on how modeling and self-efficacy influence self-regula-
tion, especially over the long term. Finally, in addition to these cognitive
benefits, the attainment of self-regulatory skill is accompanied by a grow-
ing sense of self-efficacy, which is a major source of students' intrinsic in-
terest and motivation to continue their academic learning on their own
(Bandura & Schunk, 1981; Zimmerman, 1985, 1995).
References
Bailey, T. (1993). Can youth apprenticeship thrive in the United States? Educa-
tional Researcher, 22(3), 4-10.
Bandura, A. (1988). Self-regulation of motivation and action through goal sys-
tems. In V Hamilton, G. H. Bower, & N. H. Frijda (Eds.), Cognitive perspec-
tives on emotion and motivation (pp. 37-61). Dordrecht, The Netherlands:
Kluwer.
Bandura, A. (in press). Self-efficacy: The exercise of control. New York: Freeman.
Bandura, A., & Cervone, D. (1983). Self-evaluative and self-efficacy mechanisms
governing the motivational effects of goal systems. Journal of Personality and
Social Psychology, 45, 1017-1028.
Bandura, A., & Cervone, D. (1986). Differential engagement of self-reactive in-
fluences in cognitive motivation. Organizational Behavior and Human Deci-
sion Processes, 38, 92—113.
Bandura, A., & Jeffery, R. W. (1973). Role of symbolic coding and rehearsal
Influences on self-regulation \11
processes in observational learning. Journal of Personality and Social Psychol-

ogy, 26,122-130.
Bandura, A., & Kupers, C. J. (1964). Transmission of patterns of self-reinforce-
ment through modeling. Journal of Abnormal and Social Psychology, 69, 1-9.
Bandura, A., & Schunk, D. H. (1981). Cultivating competence, self-efficacy, and
intrinsic interest through proximal self-motivation. Journal of Personality and
Bandura, A., & Walters, R. H. (1963). Social learning and personality develop-
ment. New York: Holt, Rinehart and Winston.
Berk, L. E. (1986). Relationship of elementary school children's private speech to
behavioral accompaniment to task, attention, and task performance. Develop-
Berndt, T. I, & Keefe, K. (1992). Friends' influence on adolescents' perceptions
of themselves at school. In D. H. Schunk & J. L. Meece (Eds.), Student percep-
tions in the classroom (pp. 51-73). Hillsdale, NJ: Erlbaum.
Bouffard-Bouchard, T., Parent, S., & Larivee, S. (1991). Influence of self-efficacy
on self-regulation and performance among junior and senior high-school age
students. International Journal of Behavioral Development, 14, 153-164.
Britton, B. K., & Tesser, A. (1991). Effects of time management practices on col-
lege grades. Journal of Educational Psychology, 83, 405^10.
Bruner, J. S. (1985). Models of the learner. Educational Researcher, 14(6), 5-8.
Bryan, J. H., & Walbek, N. H. (1970). Preaching and practicing generosity: Some
determinants of sharing in children. Child Development, 41, 329-354.
Cohen, E. G. (1994). Restructuring the classroom: Conditions for productive
small groups. Review of Educational Research, 64, 1-35.
Davidson, E. S., & Smith, W. P. (1982). Imitation, social comparison, and self-re-
ward. Child Development, 53, 928-932.
Diefenderfer, K. K., Holzman, T. G., & Thompson, D. N. (1985). Verbal self-mon-
itoring and solution flexibility in rule induction. Journal of Genetic Psycholo-
gy, 146, 79-88.
Dowrick, P. W (1983). Self-modeling. In P. W. Dowrick & S. J. Biggs (Eds.), Us-
ing video: Psychological and social applications (pp. 105-124). Chichester,
England: Wiley.
France-Kaatrude, A., & Smith, W P. (1985). Social comparison, task motivation,
and the development of self-evaluative standards in children. Developmental
Psychology, 21, 1080-1089.
Kanfer, F. H., & Gaelick, K. (1986). Self-management methods. In F. H. Kanfer &
A. P. Goldstein (Eds.), Helping people change: A textbook of methods (3rd ed.,
pp. 283-345). New York: Pergamon.
Locke, E. A., & Latham, G. P. (1990). A theory of goal setting and task perfor-
mance. Englewood Cliffs, NJ: Prentice-Hall.
Macan, T. H., Shahani, C, Dipboye, R. L., & Phillips, A. P. (1990). College stu-
dents' time management: Correlations with academic performance and stress.
Journal of Educational Psychology, 82, 760-768.
Mace, F. C, Belfiore, P. J., & Shea, M. C. (1989). Operant theory and research on
self-regulation. In B. J. Zimmerman & D. H. Schunk (Eds.), Self-regulated
learning and academic achievement: Theory, research, and practice (pp.
27-50). New York: Springer-Verlag.
Masters, J. C, & Santrock, J. W. (1976). Studies in the self-regulation of behavior:
Effects of contingent cognitive and affective events. Developmental Psycholo-
gy, 12, 334-348.
Meichenbaum, D. (1977). Cognitive behavior modification: A integrative ap-
proach. New York: Plenum.
Morgan, M. (1985). Self-monitoring of attained subgoals in private study. Journal
of Educational Psychology, 77, 623—630.
Mount, M. K., & Terrill, F. J. (1977). Improving examination scores through self-
monitoring. Journal of Educational Research, 71, 70-73.
Newman, R. S. (1994). Academic help seeking: A strategy of self-regulated learn-
ing. In D. H. Schunk & B. J. Zimmerman (Eds.), Self-regulation of learning
and performance: Issues and educational applications (pp. 283—301). Hills-
dale, NJ: Erlbaum.
Newman, R. S., & Schwager, M. T. (1992). Student perceptions and academic
help-seeking. In D. H. Schunk & J. L. Meece (Eds.), Student perceptions in the
classroom (pp. 123-146). Hillsdale, NJ: Erlbaum.
Rosenthal, T. L., & Bandura, A. (1978). Psychological modeling: Theory and
practice. In S. L. Garfield & A. E. Bergin (Eds.), Handbook ofpsychotherapy
and behavior change: An empirical analysis (2nd ed., pp. 621-658). New York:
Wiley.
Rosenthal, T. L., & Zimmerman, B. J. (1978). Social learning and cognition. New
Sagotsky, G., Patterson, C. J., & Lepper, M. R. (1978). Training children's self-
control: A field experiment in self-monitoring and goal-setting in the class-
room. Journal of Experimental Child Psychology, 25, 242—253.
Salomon, G. (1984). Television is "easy" and print is "tough": The differential in-
vestment of mental effort in learning as a function of perceptions and attribu-
tions. Journal of Educational Psychology, 76, 647-658.
Schunk, D. H. (1981). Modeling and attributional effects on children's achieve-
ment: A self-efficacy analysis. Journal of Educational Psychology, 73, 93-
105.
Schunk, D. H. (1983a). Developing children's self-efficacy and skills: The roles of
social comparative information and goal setting. Contemporary Educational
Schunk, D. H. (1983b). Goal difficulty and attainment information: Effects on
children's achievement behaviors. Human Learning, 2, 107-117.
Schunk, D. H. (1983c). Progress self-monitoring: Effects on children's self-effica-

cy and achievement. Journal of Experimental Education, 51, 89-93.
Schunk, D. H. (1985). Participation in goal setting: Effects on self-efficacy and
skills of learning disabled children. Journal of Special Education, 19, 307-
317.
Schunk, D. H. (1987). Peer models and children's behavioral change. Review of
Educational Research, 57, 149-174.
Schunk, D. H. (1989). Self-efficacy and achievement behaviors. Educational Psy-
chology Review, 1, 173-208.
Schunk, D. H. (in press). Goal and self-evaluative influences during children's
cognitive skill learning. American Educational Research Journal.
Schunk, D. H., & Cox, P. D. (1986). Strategy training and attributional feedback
with learning disabled students. Journal of Educational Psychology, 78,
201-209.
Schunk, D. H., & Gunn, T. P. (1986). Self-efficacy and skill development: Influ-
ence of task strategies and attributions. Journal of Educational Research, 79,
238-244.
Schunk, D. H., & Hanson, A. R. (1985). Peer models: Influence on children's self-
efficacy and achievement. Journal of Educational Psychology, 77, 313-322.
Schunk, D. H., & Hanson, A. R. (1989). Self-modeling and children's cognitive
skill learning. Journal of Educational Psychology, 81, 155-163.
Schunk, D. H., Hanson, A. R., & Cox, P. D. (1987). Peer model attributes and chil-
dren's achievement behaviors. Journal of Educational Psychology, 79, 54-61.
Schunk, D. H., & Rice, J. M. (1984). Strategy self-verbalization during remedial
listening comprehension instruction. Journal of Experimental Education, 53,
49-54.
Schunk, D. H., & Rice, J. M. (1985). Verbalization of comprehension strategies:
Effects on children's achievement outcomes. Human Learning, 4, 1-10.
Schunk, D. H., & Swartz, C. W. (1993). Goals and progress feedback: Effects on
self-efficacy and writing achievement. Contemporary Educational Psychology,
18, 337-354.
Schunk, D. H., & Zimmerman, B. J. (1994). Self-regulation of learning and per-
formance: Issues and educational applications. Hillsdale, NJ: Erlbaum.
Zimmerman, B. J. (1985). The development of "intrinsic" motivation: A social
cognitive analysis. In G. J. Whitehurst (Ed.), Annals of child development (pp.
117-160). Greenwich, CT: JAI.
Zimmerman, B. J. (1989). A social cognitive view of self-regulated academic
learning. Journal of Educational Psychology, 81, 329—339.
Zimmerman, B. J. (1990). Self-regulating academic learning and achievement:
The emergence of a social cognitive perspective. Educational Psychology Re-
view^, 173-201.
Zimmerman, B. J. (1994). Dimensions of academic self-regulation: A conceptual
framework for education. In D. H. Schunk & B. J. Zimmerman (Eds.), Self-reg-
ulation of learning and performance: Issues and educational applications (pp.

Zimmerman, B. J. (1995). Self-efficacy and educational development. In A. Ban-
dura (Ed.), Self-efficacy in changing societies (pp. 202-231). New York: Cam-
Zimmerman, B. J., & Bonner, S. (in press). A social cognitive view of strategic
learning. In C. E. Weinstein & B. L. McCombs (Eds.), Skill, will, and self-regu-
lation. Hillsdale, NJ: Erlbaum.
Zimmerman, B. J., & Kleefeld, C. (1977). Toward a theory of teaching: A social
learning view. Contemporary Educational Psychology, 2, 158-171.
Zimmerman, B. J., & Martinez-Pons, M. (1990). Student differences in self-regu-
lated learning: Relating grade, sex, and giftedness to self-efficacy and strategy
use. Journal of Educational Psychology, 82, 51—59.
Zimmerman, B. J., & Rosenthal, T. L. (1974). Observational learning of rule-gov-
erned behavior by children. Psychological Bulletin, 81, 29-42.
Zimmerman, B. J., & Schunk, D. H. (1989). Self regulated learning and academ-
ic achievement: Theory, research, and practice. New York: Springer- Verlag.
8 Social motivation: Goals and
social-cognitive processes. A comment
Carol S. Dweck
It is clear that the array of goals facing students is a daunting one. Stu-
dents are typically concerned not only with academic work, but also with
the liking and esteem of their teachers and peers - and with their parents'
reactions to their academic and social life. Even in the best of all possible
worlds, juggling this variety of concerns would be difficult. When one
considers that this is often not the best of all possible worlds - for exam-
ple, that what peers value is often not what parents and teachers value -
then coordinating the various social and academic goals becomes even
more demanding. Each of these chapters deals in an extremely thoughtful
way with the psychological factors that predict how and how well students
will confront these challenges.
Drawing on the wealth of research provided in the six chapters, I will
use a "goal analysis" to tie together the many findings and to integrate the
different perspectives represented in the chapters. I will begin by dis-
cussing the kinds of goals students may pursue in academic settings. I will
then move to an examination of the factors shown by research to affect
successful goal pursuit and school adjustment, and I will end by propos-
ing a dynamic model of goals that organizes these factors into a system of
coherent processes. As the work reviewed in these chapters attests, such a
goal analysis can be a fruitful way to understand academic and social
functioning (see Erdley; Ford; Harter; Juvonen; Kupersmidt, Buchele,
Voegler, and Sedikides; and Schunk and Zimmerman chapters; see also
Dodge, Asher, & Parkhurst, 1989; Pervin, 1982; 1989; Wentzel, 1991a;
this volume.). Indeed, we will see how certain patterns of goal pursuit ap-
pear to foster adjustment, whereas other patterns of goal pursuit appear to
undermine it.
181
What kinds of goals can students pursue in a school setting?

When we scan the array of goals available to students in a school setting,
there are, first of all, skill-related goals: those related to academic-intel-
lectual skills (and perhaps to athletic or artistic skills). In these areas, stu-
dents may be primarily concerned with proving their skills and being
judged favorably (performance goals), or with expanding their skills and
mastering new tasks (learning goals) (Dweck & Elliott, 1983; Dweck &
Leggett, 1988; Elliott & Dweck, 1988). They may also be motivated to
achieve for extrinsic reasons. For example, some students may try to do
well in school because parents promise good things if they do and threaten
bad things if they don't. (See Harter's discussion of intrinsic vs. extrinsic
motivational orientation, where the intrinsic motivational orientation in-
volves curiosity and desire for challenge - similar to learning goals -
whereas the external orientation involves a focus on such things as grades,
approval, or avoiding censure - a combination of performance goals and
extrinsic goals.)
Also of utmost importance are peer-oriented goals. Indeed, Wentzel
(this volume) presents evidence that students value social goals in school
more highly than they value learning-oriented goals. These social goals
can take many forms (see Erdley; Ford; Harter; Juvonen; and Kupersmidt,
et al. chapters; see also Wentzel, this volume.) One is liking and approval
from others, akin to performance goals in the academic domain (see Erd-
ley; Ford; Harter; Juvonen; Kupersmidt et al.). The negative side of this is
the goal of avoiding rejection, and it is most likely to be pursued when stu-
dents lose confidence in their ability to win approval (Taylor & Asher,
1984). Another social goal is to promote and develop relationships (Erd-
ley; Wentzel, this volume). This may be seen as akin to learning goals in
the academic domain in that both involve striving to develop the valued
commodity rather than to win approval for it. Also in a positive vein, some
students highly value the goal of helping their peers or facilitating their
peers' well-being (see Ford for an extensive discussion of these goals; see
also Wentzel, this volume). In contrast, some students may value the less
prosocial goal of being able to control or dominate others (and they may
value being known for this) (Erdley; see Ford's discussion of "hypermas-
culine" goals).
Some students may value personal social-moral goals, such as behav-
ing responsibly or being a sensitive, cooperative person (Ford; Wentzel,
this volume).
Others may put a high premium on instrumental goals: getting what
Goals and social-cognitive processes 183
they want when they want it or controlling valued resources. Ford also dis-
cusses a possibly related pattern: Students who pursue a hedonistic pattern
of goals, striving to maximize personal pleasure above all else.
Finally, students may pursue adult-oriented goals. As noted above, they
may strive to win the esteem of their teachers and parents, and they may
strive to do so, for example, by excelling in skill areas, in social areas, or
in their personal social-moral attributes.
This is by no means meant to be an exhaustive list of the goals students
may pursue in the school setting. Yet, when one considers this diversity of
goals, many of which are important to most children, it becomes clear that
students cannot pursue these goals one at a time or in isolation from each
other. As Dodge, Asher, and Parkhurst (1989) and Wentzel (1991a) point
out (see also Ford; Harter; Juvonen; Kupersmidt et al; and Schunk and
Zimmerman), most school situations present the opportunity to pursue
different goals simultaneously and many situations may even require stu-
dents to pursue several goals simultaneously.
For example, when one pursues academic-intellectual goals, these have
implications for adult and peer approval - and not necessarily in the same
direction. As the very interesting work by Juvonen shows, adolescents of-
ten mete out disapproval to peers who excel academically. The student
who values academic achievement is thus in the position of risking peer
rejection for pursuing academic goals. The student who courts peer accep-
tance at the expense of academic excellence risks parental sanctions, and
limits future options, by doing so.
Both Ford and Wentzel (this volume) emphasize the notion that social
and academic goals are not independent, but, rather, often have reciprocal
effects. Both stress the degree to which having prosocial goals and suc-
cessful peer relationships can foster greater engagement in school and
more positive intellectual outcomes (Kupersmidt, Coie, & Dodge, 1990;
Skinner & Belmont, 1993; Wentzel, 1991a,b). Thus it appears that optimal
school adjustment in either the social or academic arena calls for the abil-
ity to pursue and attain goals in both arenas.
As another example of the need for goal coordination, when peer con-
flicts arise, a number of goals may simultaneously come to the fore. There
may be the immediate instrumental goal of getting the thing one wants,
there may be the goal of dominating others and showing who's boss, there
may be the goal of avoiding rejection, there may be the goal of working
things out fairly, and there may be the goal of fostering the relationship.
All children may value these goals to some degree, but the task becomes
to select among them and to coordinate effectively the ones that are select-
ed (Dodge, Asher, & Parkhurst, 1989; Ford, Wentzel, Wood, Stevens, &
Siesfeld, 1989; Rabiner & Gordon, 1992; Wentzel, 1991a). I will return to
this important issue later.
What affects goal pursuit and goal persistence?

The six chapters shed considerable light on the social-cognitive factors
that influence the goals students select and the vigor with which they pur-
sue them. Below, I identify the beliefs and values that appear to be of ma-
jor importance, and following this, I elaborate upon the role these factors
can play in problems of goal pursuit, goal coordination, and adjustment.
Goal value
One obviously critical factor affecting whether children select and pursue
a particular goal is the importance of the goal and its meaning to the self.
Several of the chapters in this volume touch on this issue in various ways
(Erdley; Ford; Harter; and Schunk & Zimmerman). Erdley cites a number
of studies documenting the importance of control and dominance to ag-
gressive children (see also Crick & Ladd, 1990) and the importance of
avoiding rejection to withdrawn children (see Taylor & Asher, 1984).
Might these be the different ways in which these children protect their
self-esteem? In this vein, Harter highlights the ways in which different
goals feed into self-esteem for different students. For example, for some
students peer approval is paramount for maintaining self-esteem, whereas
for others it appears to be of far less importance.
Kupersmidt et al. make the compelling proposal that students' psycho-
logical adjustment can be understood in terms of the value they place on
various social goals or needs. Specifically, they argue that adjustment can
be predicted by the discrepancy between students' valued social needs and
their actual social relationships. When these social needs are not met, de-
pression and anxiety are likely to result (see Higgins, 1987; see also Har-
ter's chapter for a discussion of William James' contention that self-es-
teem is a function of the discrepancy between one's pretensions or
aspirations and one's actual successes).
Ford addresses not only what students do value, but what schools ought
to be teaching them to value. Here the emphasis is on making concern for
others something that one prides in oneself. The implication is that some of
the ills of our society may be linked to individualism run amok, to the em-
phasis on individuals' rights and desires over their social responsibilities.
A related factor affecting goal pursuit involves the perceived appropri-

ateness or legitimacy of the goal. Thus some children may value a particu-
lar goal, but not pursue it because they believe it is an inappropriate one
for the setting. For example, many children may value instrumental or
dominance goals (gaining valued commodities, controlling other chil-
dren), but they may not pursue them because they believe them to be inap-
propriate given their other goals, such as maintaining approval from oth-
ers or being a certain kind of person. What Erdley and Asher (1993b)
show (see also Slaby & Guerra, 1988) is that aggressive children not only
value goals that entail aggression (e.g., dominance or retaliation), but,
compared to their nonaggressive peers, they also believe that dominance
or retaliation via aggression is legitimate.
Efficacy, confidence, and perceived ability

Several of the chapters emphasize the importance of students' perceived
ability to attain their valued goals. Clearly, one will primarily adopt and
persist at goals one believes one has the ability to pursue successfully. For
example, Erdley and Asher (1993a,b) show that aggressive children have
more confidence than other children that they can successfully complete
goals involving aggression (see also Perry, Perry, & Rasmussen, 1986),
but have less confidence than other children in their ability to succeed at
prosocial goals. Personal agency beliefs play a central role in Ford's mod-
el of goal pursuit, and Schunk and Zimmerman present much compelling
evidence for the role of efficacy beliefs in the goals that students adopt
and in the quality of their self-regulatory processes as they work toward
their achievement goals. Schunk and Zimmerman also present a fascinat-
ing analysis of how children may acquire and internalize self-efficacy be-
liefs and self-regulatory competencies.
Attributions
How children interpret an event can be an important determinant of
whether they persist in their goal pursuit or whether they instead select
other goals. The work of Weiner (e.g., Weiner, 1985, 1986; Weiner & Kuk-
la, 1970), Dweck (e.g., Diener & Dweck, 1978; Dweck, 1975) and others
has demonstrated that certain attributions can lead to the abandonment of
goal striving, whereas other attributions can increase the vigor of goal
pursuit. In both intellectual-academic and social situations, children who
blame their abilities for failure and rejection tend to forsake active striving
toward success and instead focus on minimizing the failure (Diener &
Dweck, 1978; Goetz & Dweck, 1980). In contrast, those who focus on
their effort or other controllable factors tend to show more effective per-
sistence in the face of setbacks.
Attributions of intent have also been shown to play a clear role in so-
cial goal pursuit. For example, when aggressive children attribute hostile
intent to their peers (which they do even for accidental or ambiguous
acts), they appear to abandon prosocial goals and instead to adopt the goal
of retaliation (Dodge, 1980; Dodge & Frame, 1982; Erdley & Asher,
1993a).
Thus the attributions students make play an important role in goal pur-
suit. I will also argue the reverse - that the goals students are pursuing af-
fect the attributions they make. When children are overconcerned with
proving themselves intellectually or socially, they may be more likely to
see failure or rejection as an indictment of themselves (e.g., their intelli-
gence or their personality) (Elliott & Dweck, 1988). In a related way,
when children are overconcerned with dominance, control, or self-de-
fense, they may be more likely to interpret others' actions as slights or
provocations (see Erdley chapter).
In short, goals and attributions are likely to have important reciprocal
effects, and this relation will be developed further below.
Implicit theories
Students have certain self-conceptions or implicit theories that have been
shown to affect their goal choices and their goal pursuit (Dweck &
Leggett, 1988). For example, some students think of their intelligence as a
fixed entity (entity theory), whereas others think of it as a malleable qual-
ity that can be increased through their efforts (incremental theory). When
students believe that this important trait is fixed - that they have only a
certain amount of this precious commodity - they become oriented toward
proving its adequacy. Thus they tend to focus on performance goals, in
which the aim is to gain positive judgments of their intelligence.
In contrast, when students hold an incremental theory, they tend to fo-
cus more on learning goals - goals in which the aim is to increase their
abilities. That is, believing that intelligence can be increased appears to
orient students toward just that.
Moreover, Erdley, Cain, Loomis, Dumas-Hines, and Dweck (in press)
have recently shown these relations to hold in the social domain as well:
Those students who hold an entity theory of their personality (believe it is
a fixed trait) are more concerned about judgments in social settings than
those who hold an incremental theory of their personality (believe it is a
malleable quality). Moreover, as in the academic domain, such a focus on
performance goals leads to greater vulnerability to negative self-attribu-
tions and impaired persistence in the face of setbacks, such as rejection.
Thus, the implicit theories that students favor can affect the goals they
choose to pursue, as well as the persistence with which they pursue them.
As an aside, I would like to add a note on the parallels between motiva-
tional processes in the achievement and social domains. Beginning with
our work on attributions (e.g., Diener & Dweck, 1978; Goetz & Dweck,
1980) and continuing through our work on implicit theories and goals
(Dweck & Leggett, 1988; Elliott & Dweck, 1988), we have found that
every major motivational pattern we detect in the realm of achievement
has its analog in the realm of social relations. Similar self-attributions lead
to a nonpersistent, helpless response in both domains, #nd similar implicit
theories and goals appear to set up these attributions and responses in both
domains. Thus the message is that researchers should look across domains
(and across literatures) for clues to general patterns of motivation and
adaptation (see Weiner, 1986).
Knowledge of strategies for reaching goals

It is also clear that effective goal pursuit must involve knowledge of ap-
propriate strategies for pursuing a given goal or for coordinating goals
(Erdley; Ford; Juvonen; Schunk and Zimmerman). Moreover, what the ap-
propriate strategies are may change with age. Juvonen's research provides
a beautiful example of this. Young children valuing peer approval must
know that their peers approve of academic effort, but adolescents with the
same goals must know that their peers disapprove of academic effort. If
adolescents wish to maintain peer approval while maintaining academic
excellence, they must learn to hide their effort and to project the same in-
tellectual nonchalance as their classmates.
Juvonen provides another fine example of strategies children must
learn if they wish to maintain social approval. Specifically, she shows how
children learn that portraying their transgression as unintentional will mit-
igate the negative consequences of the transgression.
The work of Asher and his colleagues (Asher & Renshaw, 1981; Oden
& Asher, 1977) provides yet another example of the importance of strate-
gy knowledge in pursuing social goals. This work suggests that a problem
for many low-accepted children is their lack of strategies for interacting
with peers in ways that are rewarding to the peer. For instance, they may
not know that they should show interest in what a playmate does, or they
may not know how to cooperate on a task. Indeed, a major component of
Oden and Asher's successful intervention with unpopular children was to
coach these children to interact with peers in nonaversive ways, ways that
would maximize mutual enjoyment.
Contextual variables
In addition to the personal, social-cognitive variables that affect goal pur-
suit and attainment, it is important to mention environmental variables
that, not surprisingly, also exert major effects and can make the difference
between adjustment and nonadjustment. After all, when one speaks of ad-
justment, one is referring to how well one adapts to particular contexts.
The recent work of Eccles and her colleagues (e.g., Eccles & Midgley,
1989) points up how the environment can provide a good fit with stu-
dents' own goals for themselves, for instance, giving them opportunities
to achieve in ways they value or to develop the skills they value. Such a fit
is found to predict favorable developmental trajectories. In contrast, stu-
dents who find themselves in environments that afford them little oppor-
tunity to engage in pursuits they value are at greater risk for adjustment
problems. Both Harter and Juvonen, citing this work, discuss how adoles-
cents' educational environments, by not providing the proper fit with ado-
lescents' emerging goals and values, can foster negative self-evaluations
and negative attitudes toward learning. For example, Harter shows how
the junior high school environment can account for the observed decline
in students' intrinsic motivation to learn, and Juvonen shows how that en-
vironment can lead adolescents to adopt peer norms that devalue academ-
ic effort. In light of this, Juvonen makes the interesting suggestion that
schools might combat the peer norms against effort by decreasing compe-
tition and fostering small group cooperation, so that effort would then be
perceived as a prosocial force in the service of common achievement
goals (Ames, 1992).
Schunk and Zimmerman also speak to the importance of the child's
social environment as a medium for fostering self-efficacy beliefs and
self-regulatory competencies. For example, they present an important se-
ries of studies demonstrating how students can develop skills and gain
information about their competencies from observation of adult and peer
models.
Problems of goal choice and goal persistence

Given the variables that play a clear role in the choice of appropriate goals
and in the successful pursuit of those goals, we are now in a position to
look in greater detail at the problems that can arise as students conduct
their lives in the school setting.
One clear problem, highlighted in several of the chapters, arises when
students do not have appropriate/constructive goal priorities (Erdley,
Ford, Harter; see also Wentzel, this volume). For example, as noted by
Oden and Asher (1977), some low-accepted children do not seem to
know that having fun is one of the goals you might have when playing
with other children. Or, as Erdley points out, aggressive children appear
to value dominance, control, instrumental goals, and retaliation over re-
lationship-building goals, particularly when they feel provoked (Crick &
Ladd, 1990, Erdley & Asher, 1993a; Rabiner & Gordon, 1992). These
goal priorities do not appear to be ones that would foster rewarding and
smooth-running relationships.
Self-beliefs (e.g., efficacy beliefs, implicit theories), as noted, may
play a role in the setting of goal priorities and thus may contribute to the
problem of skewed goals. For example, problems of goal choice and goal
coordination may be fostered by students' implicit theories, particularly
by an "entity" theory, that depicts personal attributes as fixed entities
(Dweck, 1991; Dweck & Leggett, 1988). As mentioned above, the belief
in fixed traits (be it fixed intelligence or fixed personality) appears to pro-
mote a concern with how those traits will be judged. This preoccupation
with judgments can divert students from goals involving learning and de-
velopment. So, for example, those students holding an entity theory of in-
telligence have been shown to sacrifice meaningful learning opportunities
when these opportunities pose the risk of errors or negative judgments
(Dweck & Leggett, 1988). It is likely that those who hold an entity theory
of their personality would avoid risky social situations, ones that hold the
danger of rejection, even if those situations hold the possibility of devel-
oping desired relationships. Thus implicit theories may be seen as self-
conceptions that chronically affect the value that students place on impor-
tant goals. An entity theory, by making academic and social outcomes into
indices of adequacy, seems to elevate the importance of performance
goals to the detriment of learning/development goals, especially when the
threat of failure or rejection is present.
Inappropriate goals may then set up oversensitivity to certain cues, as
well as maladaptive attributions and nonconstructive responses. For exam-
pie, entering a situation with dominance and control goals may make stu-
dents hypersensitive to slights or threats to their dominance and may make
hostile attributions, anger, and retaliation more likely (see Erdley's chap-
ter). In a related vein, entering a situation with strong concerns about ac-
ceptance and rejection may make students hypersensitive to cues implying
rejection and may foster self-blame and helpless responses in the face of
rejection. In contrast, students who focus on building and maintaining re-
lationships or on increasing their skills will be more sensitive to cues that
suggest how this may best be done and will focus on varying their effort
or strategies when things go amiss, rather than assigning blame and with-
drawing or retaliating. In short, goals appear to be important organizers of
cognitive, affective, and behavioral responses, and may help us understand
why some children display strong, maladaptive reactions to seemingly
mild cues.
Goals are also fluid and dynamic, so that an event like failure, conflict,
or rejection can elicit new goals (or change the relative values of existing
goals). Thus some students may enter a situation with a full array of goals,
but may drop the appropriate ones (e.g., prosocial, relationship-maintain-
ing goals) when they confront a difficult situation. Instead, such goals as
retaliation or self-protection may come to the fore. For example, some ag-
gressive children may value prosocial, relationship-maintaining goals, but
as soon as they perceive provocation, they may relinquish these goals in
favor of goals that would reestablish their control and dominance through
retaliation. Or, children prone to social withdrawal may highly value the
goal of developing relationships, but that goal may be easily eclipsed by
the goal of avoiding rejection when social disapproval appears possible.
This problem, inappropriate abandonment of goals, may also arise
when students cannot find ways to coordinate their valued goals. For ex-
ample, as noted above, a real problem facing adolescents is that their peers
may disapprove of academic effort or achievement. Some may learn to
hide their effort or justify their achievement, but others may suppress their
effort or achievement to please peers, thus limiting the schools and profes-
sions for which they will qualify in the future. Some students, in the face
of conflict, may understand how to maintain their status and maintain the
relationship while solving the problem at hand, but others may escape into
self-protection or aggression, not because the other goals have lost their
value, but because they do not understand how to pursue all of these val-
ued goals simultaneously.
Harter brings up another interesting problem of goal coordination, and
that is the necessity of coordinating goals or selves across situations.
Some students may experience little or no anxiety over the fact that they
are somewhat different people who pursue somewhat different goals with,
say, their parents versus their peers, or their male versus female peers.
Others, however, may experience a great deal of anxiety over this, feeling
perhaps that they are false or hypocritical for doing so. In this case, stu-
dents must find ways either to maintain a more consistent set of roles and
goals across settings or to come to terms with the different roles and goals
that may be demanded in different settings.
In short, major problems can arise when students are unaware of what
goals suit the situation they are in and when they are unable to maintain
their pursuit of valued goals in the face of conflict or adversity.
A dynamic model of goals

Implicit in the above discussion is a model of goals, their causes, and their
effects. According to this model, students enter a situation with beliefs
and values that make certain goals more likely than others. Specifically,
such things as their values, their implicit theories, and their efficacy be-
liefs will affect which goals they initially pursue. These initial goals may
sensitize them to certain cues and may make certain attributions more
likely. Thus as they encounter events and outcomes in the situation, the
meaning of these events and outcomes is shaped by the goals students are
pursuing. Then, in light of the events and outcomes and the way in which
they are interpreted, students may adopt new goals for dealing with the
changed situation. This model can be nicely illustrated by means of exam-
ples we have been developing throughout - the case of aggressive children
and the case of withdrawn/helpless children.
In the case of aggressive children, these children may enter the situa-
tion with an array of goals, but the goals of dominance and control appear
to be primary among them (Boldizar, Perry, & Perry, 1989; Crick & Ladd,
1990; Erdley & Asher, 1993a; Lochman, Wayland, & White, 1993; Rabin-
er & Gordon, 1992; Renshaw & Asher, 1983). As we suggested above, this
focus on maintaining dominance and control may make them especially
sensitive to slights and provocations and may lead them to overperceive
hostile intent on the part of others. Perceiving threats to their dominance
in the form of hostile provocations, in turn, may cause them to be easily
angered and to adopt the goal of aggressive retaliation (Crick & Dodge,
1994).
In the case of children prone to withdrawal, these children may enter
social situations with an array of goals but salient among them are the
goals of gaining acceptance and avoiding rejection (social performance

goals) (See Erdley). This concern with approval and disapproval of the so-
cial self may lead these children to overperceive signs of rejection and to
attribute the perceived rejection to personal inadequacy (to make internal,
global, stable self-attributions) (Erdley et al., in press). This, in turn, may
lead them to adopt the self-protective goal of escaping from the situation
in order to minimize the possibility of further rejection.
The case of children prone to helplessness in the face of academic-in-
tellectual setbacks is a similar one. These children enter the situation with
a theory of fixed intelligence and with intellectual performance goals (the
goal of gaining positive judgments of their intelligence and avoiding neg-
ative ones) having a very high priority among their goals (Dweck &
Leggett, 1988; Elliott & Dweck, 1988). These goals make them especially
sensitive to cues signalling failure, leading them to overperceive setbacks
as failures and to attribute these failures to their lack of ability. This, in
turn, may set in motion not only a cycle of negative affect and perfor-
mance impairment, but also the tendency to adopt defensive goals and
strategies that will minimize the possibility or the meaning of further fail-
ure (such as escape from the situation or minimizing further effort) (see
Juvonen's chapter; see also Covington & Omelich, 1979).
Thus, studying the dynamics of children's goals may provide insights
into patterns of adaptive and maladaptive behavior. Indeed, it gives us a
step-by-step way of analyzing what happens as students enter situations
and experience events and outcomes in them. In such an analysis, first, we
would wish to know what hierarchy of active goals students enter a situa-
tion with; how appropriate/constructive those goals are in the situation;
and whether they have the strategies for attaining the more constructive
goals. If the more constructive goals are low on the hierarchy, we would
want to know why. Is it that children are unaware that these are the appro-
priate goals for the situation? Do they not have the confidence that they
can pursue these goals successfully? Do they have implicit theories that
foster an emphasis on particular goals at the expense of others? Have they
given up on some valued goals because they perceive conflicts among
goals and do not know how to coordinate them?
Next, we would want to know how that goal hierarchy changes as stu-
dents experience events and outcomes, such as conflicts or setbacks. Do
they maintain their full array of goals, especially the constructive ones
(perhaps even adding new constructive goals and strategies to address the
changed situation)? Or, alternatively, do they abandon the more construc-
tive goals and assume a defensive posture as they focus on such goals as
self-protection or retaliation? If the latter, why is this so? Is it because

their attributions about the self (e.g., fixed, negative self-attributions) have
caused them to lose confidence in their ability to enact the positive, con-
structive goals successfully? Or have their attributions about others (e.g.,
hostile attributions for perceived provocation) led them to feel that retalia-
tion is the appropriate goal?
As can be seen, the goal analysis provides a way to understand the
dynamics of behavior as it unfolds in a situation. It gives us a sense of
the things that can go wrong and a sense of the various ways we might
intervene to set them right. The goal analysis also gives us a way to or-
ganize the different social-cognitive variables psychologists study (effi-
cacy, attributions, implicit theories, goal and strategy knowledge) into
one framework and to begin to understand how they operate in concert
to influence students' cognitions, behavior, and affect, and ultimately,
their adjustment.
References
Ames, C. (1992). Classrooms: Goals, structures, and student motivation. Journal
of Educational Psychology, 84, 261-271.
Asher, S. R. & Renshaw, P. (1981). Children without friends: Social knowledge
and social skill training. In S. R. Asher and J. M. Gottman (Eds.) The develop-
ment of children's friendships (pp. 273-296). New York: Cambridge University
Press.
Covington, M. V & Omelich, C. L. (1979) Effort: The double-edged sword in
school achievement. Journal of Educational Psychology 77, 446-459.
Crick, N. R. & Dodge, K. A. (1994) A review and reformulation of social infor-
mation-processing mechanisms in children's social adjustment. Psychological
Bulletin, 115, 74-101.
Crick, N. R. & Ladd, G. W., (1990). Children's perceptions of the outcomes of so-
612-620.
Diener, C. & Dweck, C. S. (1978). An analysis of learned helplessness: Continu-
ous changes in performance, strategy, and achievement cognitions following
failure. Journal of Personality and Social Psychology, 36, 451^62.
Dodge, K. A. (1980). Social cognition and children's aggressive behavior. Child
Dodge, K. A., Asher, S. R., & Parkhurst, J. T. (1989). Social life as a goal coordi-
nation task. In C. Ames & R. Ames (Eds.), Research on motivation in education
(vol. 3). New York: Academic Press.
Dodge, K. A. & Frame, C. L. (1982). Social cognitive biases and deficits in ag-
gressive boys. Child Development, 53, 344-353.
Dweck, C. S. (1975). The role of expectations and attributions in the alleviation of
learned helplessness. Journal of Personality and Social Psychology, 31,
674-685.
Dweck, C. S. (1991). Self-theories and goals: Their role in motivation, personali-
ty, and development. In R. Dienstbier (Ed.), Nebraska symposium on motiva-
tion, (vol. 38. pp. 199-236). Lincoln NE: University of Nebraska Press.
Dweck, C. S. & Elliott, E. S. (1983). Achievement motivation. In P. Mussen & E.
M. Hetherington (Eds.), Handbook of Child Psychology. New York: Wiley.
Dweck, C. S. & Leggett, E. L. (1988). A social-cognitive approach to motivation
and personality. Psychological Review, 95, 256-272.
Eccles, J. & Midgley, C. (1989). Stage-environment fit: Developmentally appro-
priate classrooms for young adolescents. In C. Ames & R. Ames (Eds.), Re-
search on motivation in education (vol. 3. pp. 139-186). New York: Academic
Press.
Elliott, E. & Dweck, C. S. (1988). Goals: An approach to motivation and achieve-
ment. Journal of Personality and Social Psychology, 25, 109-116.
Erdley, C. A. & Asher, S. R. (1993a, April). To aggress or not to aggress: Social-
cognitive mediators of children s responses to ambiguous provocation. Paper
presented at the biennial meeting of the Society for Research in Child Develop-
ment, New Orleans.
Erdley, C. A. & Asher, S. R. (1993b, August). Linkages between aggression and
children's legitimacy of aggression beliefs. In S. R. Asher (Chair) Social rela-
tionships, social beliefs and aggression. Symposium presented at the annual
meeting of the American Psychological Association, Toronto.
Erdley, C. A., Cain, K. M., Loomis, C. C, Dumas-Hines, F. & Dweck, C. S. (in
press). The relations among children's social goals, implicit personality theory
and response to social failure. Developmental Psychology.
Ford, M. E., Wentzel, K. R., Wood, D. N., Stevens, E., & Siesfeld, G. A. (1989).
search, 4, 405^25.
Goetz, T. S. & Dweck, C. S. (1980). Learned helplessness in social situations.
Journal of Personality and Social Psychology, 39, 246—255.
Higgins, E. T. (1987). Self-discrepancy theory: A theory relating self and affect.
Psychological Review, 94, 319-340.
Press.

Child Psychology, 21 135-151.
Oden, S. & Asher, S. R. (1977). Coaching children in social skills for friendship
making. Child Development, 48, 495-506.
Perry, D. G., Perry, L. C, & Rasmussen, P. (1986). Cognitive social learning me-
diators of aggression. Child Development, 57, 700-711.
Pervin, L. A. (1982). The stasis and flow of behavior: Toward a theory of goals. In
M. M. Page (Ed.), Nebraska symposium on motivation: Personality-current
theory and research (pp. 1-53). Lincoln NE: University of Nebraska Press.
Pervin, L. A. (1989). Goal concepts in personality and social psychology: A his-
torical perspective. In L. A. Pervin (Ed.), Goal concepts in personality and so-
cial psychology (pp. 1-17). Hillsdale NJ: Lawrence Erlbaum Associates.
Rabiner, D. L. & Gordon, L. V (1992). The coordination of conflicting social
goals: Differences between rejected and nonrejected boys. Child Development,
63, 1344-1350.
Renshaw, P. D. & Asher, S. R. (1983). Children's goals and strategies for social in-
teraction. Merrill-Palmer Quarterly 29, 553-574.
Skinner, E. A. & Belmont, M. J. (1993). Motivation in the classroom: Reciprocal
effects of teacher behavior and student engagement across the school years.
Slaby, R. G. & Guerra, N. G. (1988). Cognitive mediators of aggression in adoles-
cent offenders: 1. Assessment. Developmental Psychology, 24, 580-588.
Taylor, A. R. & Asher, S. R. (1984). Children's goals and social competence: Indi-
vidual differences in a game-playing context. In T. Field, J. L. Roopnarine, &
M. Segal (Eds.), Friendship in normal and handicapped children (pp. 53-77)
Norwood NJ: Ablex.
Weiner, B. (1985). An attribution theory of achievement motivation and emotion.
Weiner, B. (1986). An attributional theory of motivation, New York: Springer-Ver-
lag.
Weiner, B. & Kukla, A. (1970). An attributional analysis of achievement motiva-
tion. Journal of Personality and Social Psychology, 15, 1-20.
Wentzel, K. R. (1991a). Social and academic goals at school: Motivation and
achievement in context. In M. Maehr & P. Pintrich (Eds.), Advances in motiva-
tion and achievement (vol. 7. pp. 185-212). Greenwich CT: JAI.
Wentzel, K. R. (1991b). Social competence at school: Relation between social re-
1-24.
PART TWO
Social motivation:
Perspectives on relationships
Interpersonal relationships in the school
environment and children's early school
adjustment: The role of teachers and peers
Sondra H. Birch and Gary W. Ladd
Young children face many challenges as they attempt to adjust to new

school environments, including adapting to classroom routines, perform-
ing increasingly difficult academic tasks, and negotiating the complexities
of interpersonal relationships with classmates and teachers. Although
children's success at negotiating these challenges may be affected by
many factors, most investigators have focussed on "internal" and organis-
mic characteristics of the child when attempting to account for early
school adjustment outcomes (e.g., gender, mental age, behavioral styles;
see Ladd & Kochenderfer, 1996). Further, researchers have historically
defined school adjustment in terms of children's academic progress or
achievement. For this reason, much of what we know about the precursors
of early school adjustment has come from research on the socialization of
children's cognitive skills (e.g., Lazar & Darlington, 1982).
One consequence of this research tradition is that the construct of
school adjustment has been defined rather narrowly, and the search for its
determinants has been restricted. To address these limitations, we have at-
tempted to elaborate upon earlier models of school adjustment in terms of
(a) the hypothesized precursors of school adjustment and (b) the concep-
tualization of school adjustment itself (see Ladd, 1989; 1996). As illustrat-
ed in Figure 9.1, our model is based, in part, on the assumption that suc-
cessful school adjustment originates both in the child (e.g., personality
and behavioral styles) as well as in the interpersonal environment (e.g.,
the nature of his or her relationships with parents, teachers, and peers).
Certain child characteristics undoubtedly influence children's successful
adjustment to school, and we propose that these characteristics may have
both direct and indirect effects (e.g., the quality of children's relationships
may mediate the effects of specific child characteristics on their adjust-
ment in school contexts). More importantly, we have argued that chil-
dren's interpersonal relationships may also play important roles in their
199
200 Perspectives on relationships
Child School
Characteristics Adjustment
^ "
Interpersonal
Relationships
Psychological Factors Types Perceptions

Self-perceptions School: School Liking
Social cognitions Classroom Peers Social Support
Teachers Affect
Organismic Factors Non-school: Loneliness
Gender
Parents Anxiety
Mental Age
Other Relatives Involvement
Language Ability
Neighborhood Peers Avoidance
Behavioral Factors Engagement
Contributions
Social Skills Absences
Academic Skills Support Performance
Stress Grades
Progress
Achievement
Figure 9.1. An elaborated model of early school adjustment: Hypothe-

sized precursors and adjustment outcomes.
school adjustment and progress (see Ladd, 1996; see also Harter; Hymel
et al., Juvonen, Kindermann et al., Wentzel, this volume).
Our work in this area developed from the perspective that the quality of
children's classroom peer relationships may serve either as supports or
stressors for young children as they attempt to adjust to the demands of
new school environments (e.g., Ladd & Price, 1987). Children who are
engaged in positive relationships with classmates are likely to feel more
comfortable in school and may be better able to take advantage of the so-
cial and learning opportunities they encounter in this setting. On the other
hand, children who experience peer rejection may develop negative atti-
tudes toward school that may inhibit them from further exploration and
development.
Conceptually, we have expanded this notion of relational supports and
stressors to other individuals who are in a position to influence children's
success in school (see Ladd, 1989, for further discussion of this model).
Interpersonal features 201
To this end, we have conducted a series of studies that examine how fea-
tures of the family environment are related to children's adjustment in
school contexts (e.g., Clark, 1994; Hart, Ladd, & Burleson, 1990; Ladd &
Goiter, 1988; Ladd & Hart, 1992; Profilet & Ladd, 1994). More recently,
we have begun to investigate interpersonal features of the school environ-
ment (in addition to peers) that may facilitate or inhibit children's success-
ful adjustment to school, namely the quality of the teacher-child relation-
ship (Birch & Ladd, 1994, 1996b). In this line of research, we hypothesize
broadly that having a supportive relationship with a significant adult fig-
ure in the school environment is likely to foster competent acclimation to
school, whereas a stressful teacher-child relationship may be an obstacle
to successful adjustment.
Similar developments have occurred in the field of achievement moti-
vation research, although the focus here has tended to be on academic or
educational adjustment outcomes, rather than socioemotional and other
adjustment outcomes. Whereas work in this field has often concentrated
on the individual's internal psychological characteristics (e.g., children's
motivational orientation), recent theorists have also recognized the impor-
tance of relationships among individuals. Connell and Wellborn (1991),
for example, have proposed that a sense of relatedness, or involvement
(described as the quality of children's relationships with peers and teach-
ers), may operate as a powerful motivator for children in school. Others
lend support to this notion by arguing that interpersonal factors play a ma-
jor role in promoting learning in school, and that learning may be opti-
mized in interpersonal contexts characterized by support for autonomy
and a sense of relatedness to others (Ryan & Powelson, 1991).
Both of these perspectives suggest that classroom relationships may
motivate children to more actively explore the school environment or may
inhibit children from doing so, depending on the quality of the relation-
ships they develop in this context. Although they have developed in sepa-
rate, parallel veins, both the relational approach (relationships as supports
or stressors) and the motivational approach (relationships as motivational
or inhibitory) have much to offer researchers investigating children's early
school adjustment.
In addition to elaborating our model to include interpersonal relation-
ships, we also propose that the concept of school adjustment be expanded
to address other dimensions that may define or underlie children's educa-
tional progress, including children's perceptions of the school environ-
ment (e.g., school liking), their affective experience in school (e.g., school
loneliness), their involvement or engagement in school (e.g., school
avoidance), as well as other issues related to their school performance

(e.g., task-related behaviors, academic readiness). Elsewhere we have ar-
gued that each of these dimensions is an important precursor of children's
later educational progress (Ladd, 1989, 1996; Ladd & Price, 1987). Chil-
dren with favorable school perceptions and school affect, for example, are
likely to feel more comfortable in school and therefore may be better able
to learn and benefit from their educational experiences. In addition, chil-
dren who are positively or actively involved in the school environment, or
who are not engaged in off-task behaviors, are also in a better position to
optimize the learning opportunities presented to them.
Thus, the purpose of this chapter is to consider existing empirical evi-
dence that may be consistent with the arguments above, focussing on the
relationships that children form in the school environment. Toward this
end, we will briefly discuss the types of relationships that young children
form with peers and teachers, including the strategies that researchers use
to define and identify them. Next, we will survey the theoretical rationales
that have been advanced by researchers in these areas to explain the link-
ages between interpersonal relationships and school adjustment. After
considering the basic premises that guide researchers' work in these do-
mains, we will also review the results of empirical studies that might sup-
port or refute these rationales. Finally, we will discuss, on a more theoreti-
cal level, future directions for research on both teacher-child and peer
relationships, their relative contributions to school adjustment, and behav-
ioral antecedents of these two relationship systems and subsequent school
adjustment.
Peer relationships and children's early school adjustment

Researchers have identified different types of peer relationships that chil-
dren may form in the classroom, including friendship and peer group ac-
ceptance, and have attempted to delineate various features of these rela-
tionships. While friendship is a construct that is defined at the dyadic
level, peer group acceptance uses the group as the level of analysis (see
Ladd & Kochenderfer, 1996). These peer relationship constructs, and the
methods used to operationalize them, are described in more detail below.
Friendship
In general, friendship refers to a voluntary, reciprocal relationship be-
tween two children. Bukowski and Hoza (1989) describe three different
levels of analysis for friendships: participation (i.e., the presence or ab-

sence of friends), quantity (i.e., the breadth of the friendship network),
and friendship quality (e.g., the extent to which the friendship can be char-
acterized by certain relationship features). Much of the evidence that has
been assembled on the linkages between children's friendships and their
functioning in school or child-care settings comes from elementary school
children (e.g., Ladd, 1990; Parker & Asher, 1993b) as well as preschool
children (e.g., Howes, 1988) and infants (Vandell & Mueller, 1980).
With young children, researchers often rely on interview techniques in
order to identify friendship pairs in the classroom. Children may be asked
to point to pictures of, name, or write the names of their best friends in the
classroom. Typically, a friendship is considered mutual when both mem-
bers of a dyad have chosen each other as one of their best friends. Investi-
gators have also relied on parents or teachers to identify children's friends
or validate results obtained from other friendship assessment techniques
(e.g., Ladd & Emerson, 1984). Further, with very young children, behav-
ior observations have been used to identify children who exhibit behaviors
considered characteristic of friends (e.g., spending a certain percentage of
their free time together; Gershman & Hayes, 1983; Hinde, Titmus, Easton,
&Tamplin, 1985; Howes, 1983).
Researchers have also attempted to study differences in the quality of
children's friendships, often by measuring children's perceptions of spe-
cific relationship features (see Berndt & Perry, 1986; Furman &
Buhrmester, 1985; Parker & Asher, 1993a, 1993b). Children's friendships
may differ in terms of how supportive or conflictual they are, for example,
and several researchers have developed interview procedures that are de-
signed to provide information about various facets of these relationships.
For example, Parker and Asher (1993a) developed a friendship quality in-
terview for older elementary children (adapted from Bukowski, Hoza, &
Newcomb, 1987) that yields reliable information about six different rela-
tionship features: validation and caring, help and guidance, companion-
ship and recreation, intimate exchange, conflict and betrayal, and conflict
resolution. Many of these same relationship features have also been reli-
ably assessed with young children (Ladd, Kochenderfer, & Coleman, in
press).
A number of propositions have been advanced about the role that
friendships play in children's school adjustment. Building on research
suggesting that friendships yield psychological benefits for children (e.g.,
Berndt, 1989; Bukowski & Hoza, 1989; Parker & Asher, 1993b), Ladd
and his colleagues (e.g., Birch & Ladd, 1994; Coleman & Ladd, 1993;
Ladd, 1990; Ladd & Kochenderfer, 1996) suggest that friendship (and
peer group acceptance in general) acts as a support for young children in
the school environment, and may therefore help children acclimate to the
school setting. Children who have a mutual friend in the classroom may
be able to use that individual as a source of emotional or instrumental sup-
port or perhaps as a secure base from which they can explore the school
environment (see Howes, 1988). The mere presence of or participation in
a friendship with a classmate may act as a protective factor for children
who might otherwise be at risk for negative school experiences (e.g., feel-
ings of loneliness).
Findings from several recent studies can be interpreted as support for
the assertion that friendships serve a supportive function for children in
the school environment. Ladd (1990) found that, following the transition
from preschool to grade school, kindergarten children who formed new
friendships with classroom peers performed better academically than did
other children, and children with friends had more positive school atti-
tudes by the second month of school than did children without friends.
Others have found that children without classroom friends were more
likely to feel lonely in school than were children with friends (Parker &
Asher, 1993b). Finally, Parker and Asher (1993a) report that rejected chil-
dren with a friendship in the classroom felt less lonely in school than did
rejected children without a classroom friendship. These findings support
the view that friendship may serve a protective function for certain groups
of children considered vulnerable to school adjustment difficulties.
Further, participation in friendship (i.e., the presence or absence of a
friend) and friendship quality may make separate contributions to chil-
dren's school adjustment. Parker and Asher (1993b) report that friendship
(participation) and friendship quality made unique contributions to the
prediction of loneliness in school. They found that, among children who
had classroom friends, children who rated their friendships as lower on
positive qualities and higher on negative qualities reported feeling more
lonely in school than did children with friendships rated more positively.
Related findings were obtained by Ladd, Kochenderfer, and Coleman (in
press) in a study that explored the relation between the quality of kinder-
garten-age children's classroom friendships and their school adjustment.
These investigators found that children who saw their friends as sources of
validation and aid were more likely to develop positive perceptions of
their classmates and favorable attitudes toward school. Further, among
boys, classroom friendships that were characterized by high levels of con-
flict were associated with multiple forms of school maladjustment, in-
cluding higher levels of school loneliness and avoidance, and lower levels
of school liking and engagement. The results of these studies highlight the
importance of examining the separate contributions of friendship partici-
pation and friendship quality to children's early school adjustment.
Peer group acceptance

Whereas children's friendships with classmates are dyadic, reciprocal, and
voluntary, children's peer group acceptance is an index of how well chil-
dren fit into the social network of the classroom. Unlike friendships, peer
group acceptance is analyzed at the classroom level, and consists of uni-
lateral nominations or ratings. Peer group acceptance is typically assessed
by asking children to rate how much they like (or like to play/work with)
each of their classmates (see Asher, Singleton, Tinsley, & Hymel, 1979).
An average rating is computed for each child in the classroom; children
who recieve higher average ratings are defined as well-accepted by their
classmates. Some researchers examine children's social standing in the
classroom by forming sociometric status groups (e.g., Coie, Dodge, &
Coppotelli, 1982). According to this procedure, children are asked to
nominate classmates that they like and don't like (or like to play/work
with and don't like to play/work with). From these nominations, social
preference (number of positive nominations minus number of negative
nominations) and social impact (positive plus negative nominations)
scores are computed. Finally, separate status groups are formed that indi-
cate varying types of acceptance and rejection by classmates: popular, ne-
glected, rejected, average, and controversial. Researchers establish various
cut-off scores for inclusion in particular groups; however, the general na-
ture of each sociometric status group remains the same. Specifically, pop-
ular children are children who are well-liked by many of their classmates,
and disliked by few. Rejected children, on the other hand, are highly dis-
liked and not positively regarded by many of their classmates. In contrast,
neglected children are neither liked nor disliked by their classmates (i.e.,
their social impact is relatively low). Controversial children receive many
positive and negative nominations, indicating that they are liked by some
children and disliked by others.
Similar to the rationale proposed for friendships (e.g., Ladd, 1990), ac-
ceptance by the peer group may function as a support for young children.
Acceptance by one's peer group may provide young children with a sense
of inclusion or belongingness in the classroom, and thus facilitate explo-
ration of and adjustment in this context (including future social relation-
ships). On the other hand, being rejected by one's classmates may operate
as a stressor, and be detrimental to children's successful adjustment to
school. It is conceivable that children who are rejected by their peers de-
velop negative "working models" of relationships that lead them to expect
further rejection and discourage them from exploring potentially positive
relationships with others (see Bowlby, 1982). Indeed, there is evidence
that negative expectations for interpersonal relationships are associated
with peer rejection (Rabiner & Coie, 1989). Peer rejection may also inhib-
it motivation to behave appropriately or to learn the academic and social
skills necessary to successfully adapt to school. Poor peer group accep-
tance may foster negative feelings toward school and the schooling experi-
ence, and may result in children withdrawing from both academic and so-
cial learning situations. Additionally, children rejected by their peers may
display frustration with their sociometric status by acting aggressively to-
wards their classmates or by being disruptive during learning situations,
which may perpetuate their maladjustment to school (see Coie, 1990, for
further discussion of this issue).
Peer group acceptance and rejection have been linked to various as-
pects of young children's school adjustment, including their academic per-
formance (e.g., Green, Forehand, Beck, & Vosk, 1980; Ladd, 1990), their
school affect and attitudes (Birch & Ladd, 1994; Ladd, 1990; Parker &
Asher, 1993a, 1993b), and their school avoidance (Birch & Ladd, 1994;
Ladd, 1990). These studies reveal that children who are relatively well-ac-
cepted by their classmates tend to demonstrate better school adjustment
than do their less well-accepted peers. Ladd (1990), for example, found
that rejection by the peer group was predictive of negative perceptions of
school and greater school avoidance. These findings support the assertion
that poor peer relationships may act as stressors for children and restrict
their ability to cope with the demands of school.
Studies in which children are classified into sociometric status groups
yield similar results, indicating that children who are rejected by their
classmates exhibit poorer school adjustment than do children who are not
rejected (e.g., Cassidy & Asher, 1989; Coie & Dodge, 1988; Crick &
Ladd, 1993; Ladd, 1990). Ladd (1990) found that rejected kindergarten
children developed less positive perceptions of school, displayed higher
levels of school avoidance, and exhibited lower levels of school perfor-
mance than did popular, average, or neglected children. Further, Cassidy
and Asher (1992) reported that first-grade children who were rejected by
their peers felt more lonely in school than did children in other sociomet-
ric status groups. Crick and Ladd (1993) report similar results for third-
and fifth-grade children. Rejected status has been found to be related to
academic problems (Coie & Dodge, 1988) as well. Together, these studies
further highlight the role of sociometrically-defined peer rejection as a

potential risk factor, or precursor of young children's school adjustment
difficulties.
Another form of peer rejection that may operate in the school environ-
ment and lead to poor adjustment outcomes is peer victimization. Chil-
dren who are bullied by their classmates, or targeted for various forms of
peer aggression, may view school as an unsafe or threatening place. They
may thus develop negative feelings about this environment and seek to
avoid it. Further, it is likely that these negative perceptions influence chil-
dren's later adjustment; indeed, some have argued that children must per-
ceive their environment as safe in order to successfully meet the academic
demands of school (Hoover & Hazier, 1991). In addition, the victimized
child may find it difficult or undesirable to become actively engaged in ei-
ther academic or social classroom activities. A recent study designed to
investigate links between peer victimization and children's school adjust-
ment lends support to these assertions (Kochenderfer & Ladd, in press). In
this study, peer victimization was positively associated with kindergarten
children's reports of loneliness and school avoidance, and was negatively
assoicated with their reports of school liking. These findings emphasize
the importance of studying different forms of peer rejection in order to
clarify the concomitants and precursors of children's early school adjust-
ment difficulties.
The teacher-child relationship and children's

early school adjustment
Classroom teachers are also significant figures in the school environment
with whom children can form relationships, and these relationships may
have important consequences for children's school adjustment (see also
Harter; Wentzel, this volume). The teacher-child relationship has primari-
ly been described from an attachment perspective (e.g., Howes & Hamil-
tion, 1992, 1993; Howes & Matheson, 1992), although some researchers
have recently used a relationship features perspective as a framework for
studying this relationship (e.g., Birch & Ladd, in 1994, 1996a, 1996b), as
described below.
The teacher-child relationship as an attachment relationship

Several researchers who have studied the ties that develop between teach-
ers and children have used the principles found in attachment theory to
characterize these relationships. Howes and her colleagues (Howes &

Hamilton, 1992, 1993; Howes & Matheson, 1992) rely on principal con-
structs from the literature on parent-child attachment relationships to de-
scribe the teacher-child relationship. In one study, for example, Howes
and Hamilton (1993) identified secure, avoidant, and ambivalent children
using observers' ratings of children's behaviors. These teacher-child at-
tachment classifications were based on the parent-child attachment classi-
fications defined by Ainsworth and her colleagues (Ainsworth, Waters, &
Wall, 1978). Because of the use of the attachment perspective in this area
of research, many of the examinations of the teacher-child relationship
have focussed on children in preschool and kindergarten (cf. Lynch & Ci-
cchetti, 1992).
Attachment theory has also been used by other researchers who have
attempted to characterize teacher-child relationships, although the prima-
ry constructs are often not explicitly linked to the parent-child attachment
literature (e.g., Lynch & Cicchetti, 1992; Pianta & Steinberg, 1992). Em-
ploying the construct of teacher-child relatedness, Lynch and Cicchetti
(1992) describe five patterns of teacher-child relationships that vary in
emotional quality and psychological proximity seeking: optimal, de-
prived, disengaged, confused, and average. These patterns were based on
children's reports of their relationships with their teachers.
Alternatively, Pianta, Steinberg, and Rollins (1995) have examined fea-
tures of the teacher-child relationship using teachers' perceptions of their
relationships with their students. The quality of the teacher-child relation-
ship is assessed via the Student-Teacher Relationship Scale (STRS),
which is comprised of items derived from attachment theory and research
on teacher-child interactions. Items were designed to tap dimensions of
warmth/security, anger/dependence, and anxiety/insecurity - dimensions
reminiscent of the constructs found in the parent-child attachment litera-
ture. Recent analyses of the STRS have revealed three distinct factors:
closeness, dependency, and conflict.
Despite these efforts to describe the teacher-child relationship, little at-
tention has been paid as to how this relationship might be associated with
children's adjustment in school contexts. Howes and Hamilton (1992)
note that one of the various roles or functions of young children's teachers
is that of a caregiver who is responsible for the physical and emotional
well-being of the child in the absence of the parent. By providing a secure
base from which children can explore their surroundings, teachers may fa-
cilitate children's successful adaptation to the school environment. Also
consistent with the attachment perspective is the notion that children may
form internal working models of relationships based on the feedback that
they receive during teacher-child interactions (see Bowlby, 1982; Skinner

& Belmont, 1993). These models of relationships may, in turn, influence
children's ability to successfully adjust to school, both socioemotionally
and academically.
In one of the few studies to directly address the relation between
teacher-child relationships and children's school adjustment, Pianta and
Steinberg (1992) found that children who were recommended for reten-
tion in kindergarten, but who were actually not retained, had more positive
relationships with their teachers than did those children who actually were
retained. The use of a single index of school adjustment (e.g., retention),
however, implies a relatively narrow view of school adjustment; a more
comprehensive assessment of school adjustment is needed in future stud-
ies (see Birch & Ladd).
The teacher-child relationship as conceived within a relationship

features perspective
Another conceptual framework that has been used to guide research on the
teacher-child relationship is the relationship features perspective (see
Birch & Ladd, 1994, 1996a, 1996b). This perspective has its origins in
past social-psychological theory and research (see Weiss, 1974), and has
recently been applied to the study of children's interpersonal relationships,
especially friendships (see Berndt & Perry, 1986; Furman & Buhrmester,
1985; Parker & Asher, 1993a). Researchers who employ a relationship
features perspective seek to delineate the properties of relationships (e.g.,
interactional processes) that create specific "provisions" for the individu-
als who participate in the relationship (e.g., psychological costs or bene-
fits). Thus, relationship features can be conceptualized as dynamic
processes (e.g., observable interactions) or felt properties of the relation-
ship (e.g., perceived attributes of qualities) that may facilitate certain out-
comes for the individual participants (e.g., sense of well-being, adjust-
ment).
Although teachers and children undoubtedly form relationships that
possess a myriad of features, our aim has been to identify those that make
important contributions to young children's early school adjustment. As
developmentalists, we recognize that the nature of children's relationships
and the provisions they yield change over time. Thus, certain relationship
features may emerge or become more or less important depending on the
age of the child, and the types of environmental and developmental tasks
the child encounters during corresponding time periods. For example,
whereas certain features of the teacher-child relationship may be especial-
ly important for young children, other features may emerge over time and
have greater adaptive significance as children grow older or the demands
of the school environment change. Drawing upon work by Pianta, Stein-
berg, and Rollins (in press) and others, we have proposed three distinct
features of the teacher-child relationship that are particularly important
for young children, especially as they negotiate the transition to grade
school: closeness, dependency, and conflict (Birch & Ladd, 1996b).
Closeness, as a feature of the teacher-child relationship, is reflected in
the degree of warmth and open communication that is manifested between
a teacher and a child, and it may function as a support for young children
in the school environment. Indeed, our conception of closeness encom-
passes the extent to which children seem comfortable approaching the
teacher, talking about their feelings and experiences, and using the teacher
as a source of support or comfort when upset. Having warm and open
communication with a significant figure in the classroom (e.g., the
teacher) may facilitate positive affect and attitudes towards school. In ad-
dition, supportive teacher-child relationships may motivate children to be-
come more involved or engaged in the school environment. In this man-
ner, closeness may also encourage young children's learning and
performance in school.
Contrasted with closeness, dependency can be construed as a rela-
tionship quality that interferes with children's successful adjustment to
school. Attachment theorists have distinguished between attachment
(which has positive connotations) and dependency (which has negative
developmental implications; see Bowlby, 1982). Optimally, in supportive
relationships, it is considered adaptive for closeness to increase over time
and for dependency to decrease over time. As conceptualized in our
work, dependency refers to possessive and "clingy" child behaviors that
are indicative of an over-reliance on the teacher. A particular teacher-
child relationship may be characterized as very close without being high-
ly dependent as well. Likewise, some children may be quite dependent
on their teachers without sharing a close relationship with them. Thus,
the two constructs may make distinct contributions to children's early
school adjustment.
Children who are overly dependent on the classroom teacher may be
tentative in their explorations of the school environment; they may feel
less motivated to explore their surroundings or other social relationships.
By spending an inordinate amount of time with the teacher, children may
not be engaging in learning or social tasks in the classroom. Feelings of
loneliness and anxiety, as well as negative feelings about and attitudes to-
ward school, may also be more common in children who display higher
levels of dependency on the teacher.
Finally, it is likely that conflict in the teacher-child relationship func-
tions as a stressor for children in the school environment, and may impair
successful adjustment to school. Conflictual teacher-child relationships
are characterized by discordant interactions and a lack of rapport between
teacher and child. Children experiencing conflict in their relationships
with classroom teachers may be limiting the extent to which they can rely
on that relationship as a source of support. As a potential stressor in the
school environment, teacher-child conflict may be emotionally upsetting
to young children, yielding (or perpetuating) a variety of negative emo-
tions (e.g., anger, anxiety) and behaviors (e.g., aggression, noncompli-
ance). In this manner, such friction in the teacher-child relationship may
be antithetical to motivating children to become positively engaged in the
school environment. A conflictual teacher-child relationship may be relat-
ed to children becoming disengaged or uninvolved or may foster feelings
of alienation and loneliness, in addition to negative school attitudes. An
aversive teacher-child relationship may be associated with academic prob-
lems as well.
In a study designed to address these issues, we found that each of these
features of the teacher-child relationship was related to various aspects of
kindergarten children's school adjustment (Birch & Ladd, 1996b). Teach-
ers rated children's school perceptions (school liking), engagement in the
school environment (e.g., school avoidance, self-directedness, cooperative
participation), as well as the quality of their teacher-child relationships.
Children's school attitudes, as reported by teachers, were correlated with
all three teacher-child relationship features; specifically, children with rel-
atively close, nondependent, and nonconflictual relationships with their
teachers were seen as liking school more than were children with less pos-
itive relationships with their teachers. In addition, children's involvement
in the school environment was also related to the quality of their teacher-
child relationships. Higher dependency and conflict were both associated
with greater school avoidance. Further, all three teacher-child relationship
dimensions were linked with ratings of children's self-directedness, indi-
cating that relatively high closeness, low dependency, and low conflict
were related to more self-directed functioning in the classroom. Low con-
flict was also associated with higher ratings of obedience in the classroom
(i.e., adapting to classroom routines). Finally, children's self-reports of
school avoidance were positively related to greater dependence on the
teacher.
In a related vein with older children, Wentzel (1993; this volume)

found that teacher preference (i.e., teacher ratings of how much he or she
would like to have the student in class again next year) was positively re-
lated to academic performance outcomes in sixth- and seventh-graders. In
addition, Skinner and Belmont (1993) reported that 8-to 12-year-old chil-
dren who had intimate relationships with their teachers had more positive
school affect and attitudes in the classroom. These authors noted that chil-
dren's working models of teacher-child relationships may be influenced
by the quality of these relationships, which may influence their subse-
quent affective experience and engagement in the school environment.
Relationships and adjustment: Prospective theory

and research
In light of the theory and research presented above, it is evident that there
are a myriad of directions for future investigations of children's interper-
sonal relationships and their adjustment in school contexts. The discus-
sion that follows highlights key issues that may be relevant for investiga-
tors who wish to further pursue these questions, including: (a) linkages
between peer and teacher-child relationships, (b) relative contributions of
peer and teacher-child relationships to children's early school adjustment,
and (c) the role of children's behavior in forming relationships and adjust-
ing to school.
Linkages between peer and teacher-child relationships

Although some researchers have alluded to the association between chil-
dren's relationships with both teachers and peers, few have explicitly ad-
dressed the nature of this relation. Some research suggests that children
experiencing difficulties with peers experience higher rates of interaction
with their teachers (e.g., Asarnow, 1983; Dodge, Coie, & Brakke, 1982;
Goldman, Corsini, & deUrioste, 1980; Howes, Phillips, & Whitebrook,
1992; Marshall & McCandless, 1957). Other investigators have found no
connections between these two relational systems (e.g., Moore & Upde-
graff, 1964).
One of the few studies to explicitly address this issue (Howes, Mathe-
son, & Hamilton, 1994) found that four-year-old children classified as
having secure relationships with their teachers (as rated by observers)
were better accepted by their peers than were children who had insecure
teacher-child relationships. This relation was also explored by Birch and
Ladd (1994) who found that children with relatively low conflict, low de-
pendency, or high closeness in their teacher-child relationships were better

accepted by their classroom peers than were children with high conflict,
high dependency, or low closeness.
As noted above, closeness in the teacher-child relationship refers to the
degree of warmth and open communication present in the relationship.
Similarly, peers' sociometric ratings indicate the extent to which they want
to affiliate with each of their classmates, and these ratings are also likely
to be related to engaging, positive behavioral styles (characterized, for ex-
ample, by warmth, open communication, friendly initiations, and so on).
Thus, the two relationship systems may be positively related as a result of
certain personality or behavioral characteristivs of the child (e.g., maturi-
ty, warmth, prosocial behavior) that elicit positive feelings from both
teachers and peers. Alternatively, perhaps children who form close rela-
tionships with their teachers are able to use her or him as a secure base
from which they can explore the classroom environment; this exploration
may include the formation of social relationships with their classmates.
Dependency in the teacher-child relationship may be negatively related
to competent functioning in the peer group, because children who are
highly dependent on the teacher may be relatively immature or lack the
social skills necessary to pursue successful relationships with their class-
mates. They also may be children who are relatively adult-focussed, and
more interested in adult (rather than peer) attention. Because there is the
expectation for children's dependency to diminish over time, children who
remain highly dependent on the classroom teacher may be violating class-
room norms for behavior. It is possible that children who seem highly de-
pendent on the teacher are perceived as immature by the peer group, and
thus may be less well-accepted than their more independent peers. In ad-
dition, dependency on the teacher may result in children being less avail-
able for participation in peer activities, thus producing and maintaining
their social status in the classroom. It is also plausible that children who
are socially withdrawn or experience impaired peer relationships might
become overly dependent on the teacher as a source of validation and sup-
port.
Conflict may be present in both peer and teacher-child relationships,
because some children may have an aversive style of interaction that oper-
ates across various interpersonal relationships and negatively influences
the quality of relationships in both domains. Alternatively, children might
construct negative working models of relationships (see Bowlby, 1982) or
develop negative expectations for relationships (see Rabiner & Coie,
1989) that serve to frame all interpersonal interactions in a negative light.
According to Bowlby (1982), children's working models develop from
their behaviors and the feedback that they receive for these behaviors. A
child exhibiting behaviors considered aversive to both teachers and peers
(e.g., aggression) might experience both rejection by the peer group and a
conflictual relationship with the teacher, which may serve to reinforce
negative expectations for future relationships with others. This might re-
sult in a conflictual style of interaction that is counter-productive in terms
of forming or maintaining positive relationships with teachers and peers.
Interpersonal features of the school environment and children s

early school adjustment
Currently, there is paucity of systematic investigations into the relative
contributions of these two relationship systems (peer and teacher-child) to
children's early school adjustment. A preliminary study designed to ad-
dress this issue yielded promising results (Birch & Ladd, 1994). In this
study, the quality of children's relationships with both classroom peers
and teachers emerged as important correlates of children's early adjust-
ment to school. For example, consistent with previous research (e.g.,
Parker & Asher, 1993b), we found that children with higher quality peer
relationships reported feeling less lonely in school. Beyond this linkage,
however, our findings also revealed that conflict in the teacher-child rela-
tionship was positively correlated with children's feelings of loneliness in
school. Moreover, the degree of conflict in the teacher-child relationship
emerged as a significant predictor of changes in loneliness from fall to
spring of the school year.
Our findings also suggest that conflict and dependency in the teacher-
child relationship may be relevant to young children as they form their ini-
tial school attitudes (e.g., school liking). Children with highly dependent
or conflictual teacher-child relationships reported liking school less than
did their classmates with less dependent or conflictual teacher-child rela-
tionships. Peer group acceptance was not as strong a predictor of school
liking as was teacher-child conflict, suggesting the need for further inves-
tigation of the relative strengths of the impact that these two relationship
systems have on children's adaptation in school contexts.
Finally, children with positive relationships in both teacher-child and
peer domains expressed less desire to avoid school than did children with
poor relationships in both domains. The level of teacher-child conflict in
the fall was further related to children's school avoidance desires in the
spring semester, indicating that early discord in the teacher-child relation-
ship continued to be associated with children's desire to avoid the school
environment later in the school year. These results support the assertion
that it is important to study the quality of children's relationships with
both teachers and peers when considering how well they will adjust to the
school environment.
These preliminary findings are consistent with the premise that both
classmates and teachers have the potential to serve as emotional and in-
strumental supports for young children in the school environment. Posi-
tive relationships with significant figures in school are likely to promote
successful adaptation to school, whereas contentious relationships with
these figures may discourage effective school adjustment. Children who
have positive relationships with both teachers and classmates may demon-
strate better adjustment to school than those children who have poor rela-
tionships in both domains. As Connell and Wellborn (1991) have pointed
out, the quality of these relationships may serve an important motivational
function for children. Future studies should more explicitly examine the
mechanisms by which these interpersonal features of the school environ-
ment might motivate and encourage children to successfully adapt to the
school environment.
What is of greater interest, perhaps, is the type of adjustment demon-
strated by children with discordance in terms of the quality of relation-
ships in these two systems (e.g., good teacher-child and poor peer rela-
tionships or poor teacher-child and good peer relationships). It is possible
that teacher-child and peer relationships may make compensatory contri-
butions to various aspects of children's school adjustment. Positive peer
relationships, for example, might buffer children with poor teacher-child
relationships against disliking school or feeling the desire to avoid the
school environment. Similarly, a close relationship with the teacher might
serve as a protective factor against loneliness for children experiencing
peer relationship difficulties (e.g., low acceptance, lack of a best friend).
It has been shown that having even one supportive, close relationship with
another person in the school environment (e.g., a best friend) can buffer
children against negative school adjustment outcomes (see Parker & Ash-
er, 1993a). Further research in this area is necessary to determine whether
the benefits that accrue from being involved in positive relationships in
one domain outweigh the potential for negative school adjustment out-
comes that might result from having less than optimal relationships in the
other domain.
Although there is some degree of overlap in terms of the functions of
the relationships that young children form with peers and teachers, there
are also distinctive functions that these two relationship systems serve.
According to Hartup (1989), children are involved in social relationships

that can be characterized as either vertical (relationships with individuals
who have greater knowledge and social power; e.g., teacher-child relation-
ships) or horizontal (relationships with individuals who possess the same
social power, characterized by reciprocity and expectations of equality;
e.g., peer relationships). The functions of vertical relationships include
providing protection and security (on the part of the adult), as well as
modeling basic social skills. In horizontal relationships, social skills are
elaborated, and themes of cooperation and competition emerge.
Thus, perhaps the quality of teacher-child relationships forecasts dif-
ferent adjustment outcomes than does the quality of children's peer rela-
tionships. Children may turn to significant figures in the school environ-
ment for different types of support (e.g., emotional, instrumental,
academic, social) and, therefore, difficulties in one relationship domain
may be reflected in some areas of school adjustment, but not others. Fric-
tion in the teacher-child relationship, for example, may be linked with
poor academic performance, whereas peer rejection may be associated
with feelings of loneliness in school, especially as children approach ado-
lescence and the peer group has an increasing impact on children's devel-
opment and adjustment. Other issues include consideration of whether
school adjustment outcomes such as school liking, for example, are more
closely tied to specific features of the teacher-child relationship (e.g.,
closeness) or acceptance by the peer group. In a related vein, future re-
search should consider whether teacher-child and peer relationships make
unique contributions to the same adjustment outcomes; that is, are their
effects on school adjustment additive (rather than compensatory, as dis-
cussed above)? Certain aspects of children's school liking, for example,
may be associated with the quality of their peer relationships (e.g., liking
social activities and games), whereas other facets may be related to the
quality of the teacher-child relationship (e.g., enjoying learning activities).
Issues such as these have yet to be investigated; future research will help
to illuminate the answers to these question.
Examining the behavioral antecedents of peer and

teacher-child relationships
An extensive body of literature addresses the behavioral antecedents and
correlates of children's peer relationships. Aggressive behavior (e.g.,
Dodge, 1983; Dodge, Coie, & Brakke, 1982) in particular has been linked
to rejection by the peer group. At present, there is a substantial body of ev-
idence indication that aggression is an important antecedent of poor peer

relationships in preschool, as well as elementary school samples (see
Coie, Dodge, & Kupersmidt, 1990, for a review). The link between with-
drawal and peer rejection is not as clear, although some studies suggest
that withdrawing from the peer group is negatively related to peer group
acceptance, especially for girls (e.g., Cantrell & Prinz, 1985; La Greca,
1981). For example, Ladd (1983) found that rejected girls were three
times as likely to engage in parallel play than were nonrejected girls or
boys. Moreover, in preschool and kindergarten, withdrawn children exhib-
it less mature forms of play and are the recipients of fewer social initia-
tions from peers than are their classmates (Rubin, 1982). As children get
older, sociometric status is also associated with more subtle behaviors, in-
cluding giving and receiving help, sharing, group entry skill, and honesty
(Carlson, Lahey, & Neeper, 1984).
Compared to the body of evidence in the peer domain, the research
identifying the behaviors associated with different types of teacher-child
relationships is less clear. Wentzel (1991) notes that teachers prefer stu-
dents who exhibit certain types of behaviors (e.g., cooperative, cautious,
responsible) but not others (e.g., independent, assertive, disruptive; see
Wentzel, 1991 for further discussion of this issue). Some researchers have
used behavior observations in order to classify the types of relationships
that children have with their teachers. Using the Waters and Deane Attach-
ment Q-Sort, Howes and Hamilton (1993) identified secure, avoidant, and
ambivalent/resistant children according to observers' ratings of their be-
haviors during teacher-child interactions. Secure children were those who
scored high on many of the following behavioral indices: predominantly
happy, easily comforted, comfort soliciting, and obedient. Avoidant chil-
dren, in contrast, scored high on ratings of being unaware of changes in
caregivers, expecting adult to be unresponsive, and demanding initiation.
Finally, ambivalent children were seen as highly demanding and impa-
tient, having distressed social interaction, and crying often, as well as be-
ing low on physical contact. It should be noted, however, that this study
used behavioral observations in order to classify children into various
teacher-child relationship categories (i.e., the relationship classification
was determined by the children's exhibited behaviors). In this way, it dif-
fers from other studies of the behavioral correlates of teacher-child rela-
tionships, in which the quality of the teacher-child relationship is assessed
using measures that are separate from the behavioral indices (e.g.,
Wentzel, 1993).
It is likely that some of the same behaviors associated with peer rela-
tionship difficulties are similarly associated with problematic teacher-

child relationships. Aggressive and disruptive behavior, for example, may
be indicative of a style of interaction that is undesirable to both teachers
and peers. Indeed, Safran and Safran (1985) report that teachers find ag-
gressive behavior to be the most disruptive to classroom order. In addi-
tion, certain types of withdrawn behavior may be characteristic of children
who isolate themselves not only from the peer group, but from contact
with the teacher as well (e.g., behavioral inhibition).
On the other hand, there may be some behavioral styles that are con-
ducive to forming and maintaining positive relationships in one domain,
but promote negative relationships in the other domain. Behaviors that
lead children to become "teachers' pets," for example, might be inimical
to peer group acceptance or lead to resentment by the peer group (e.g.,
compliance that may be perceived as submissive behavior by peers). Mak-
ing peers laugh might be associated with peer group acceptance, but may
be problematic for teachers if this behavior is disruptive and interferes
with learning activities. Indeed, such behaviors seem to have inconsistent
outcomes even within the peer culture; behaviors such as being funny and
making peers laugh are related to controversial sociometric status (i.e.,
children who are liked by a number of children and also disliked by a
number of children; Coie, Finn, & Krehbiel, 1984). Further, as children
move toward adolescence, teachers (and other adults) and peers may have
increasingly divergent behavioral expectations, and children's ability to
successfully negotiate these competing demands may be challenged (see
also Juvonen, this volume). Although the influence of teachers may be
more significant to young children, the peer group may become increas-
ingly influential as children grow older. Some investigators have started to
examine how children coordinate multiple social and academic goals, and
how this goal pursuit is related to their acceptance by teachers and peers
(e.g., Wentzel, 1994); however, further discussion of the goal coordination
literature is beyond the scope of the present chapter (see Ford; Wentzel,
this volume).
We recently examined how children's early behavioral styles are relat-
ed to the types of relationships that they form with classroom teachers and
peers (Birch & Ladd, 1996a) both concurrently (i.e., fall of kindergarten)
as well as at two later points in time (i.e., spring of kindergarten and
spring of first grade). At all three assessment times, aggression and hyper-
activity were associated with problematic relationships with teachers and
peers, and prosocial behavior was linked to better relationships in both do-
mains. Asocial behavior was negatively related to children's peer group
acceptance, and was associated with poorer quality teacher-child relation-
ships (especially for girls). Finally, anxious-fearful behavior was not relat-
ed to acceptance by the peer group, but was negatively associated with
harmonious teacher-child relationships (particularly for girls). These find-
ings suggest that the behaviors that children exhibit early in kindergarten
have enduring relationship outcomes, even with new teachers and peers
(in first grade). Additionally, many of the same behaviors that are linked
to the level of children's acceptance by the peer group were also associat-
ed with the quality of the relationships that they have with their classroom
teachers. These data also suggest that there are important gender differ-
ences in the behaviors that are related to children's interpersonal relation-
ships in the school environment, and these differences warrant further in-
vestigation.
Finally, there is also the need for a comprehensive examination of the
links between classroom behavior, relationships with teachers and peers,
and adjustment in school contexts. Researchers have found early aggres-
sive and disruptive behavior in elementary school to be associated with
school adjustment difficulties in high school, including poor grades (Feld-
husen, Thurston, & Benning, 1970) and dropping out of school (see Park-
er & Asher, 1987 for a comprehensive review). Other correlational studies
have found appropriate behaviors to be associated with positive academic
outcomes (see Wentzel, 1991).
We propose that behavior may be associated with young children's ear-
ly school adjustment in at least two ways. First, behavior may be directly
linked to how well children respond to the demands of the school environ-
ment. Specifically, it is possible that certain behaviors have direct effects
on children's school adjustment (e.g., aggressive behavior may lead to
children being removed from learning and social situations, which may re-
sult in impaired academic performance and more negative school atti-
tudes). Second, behavior may be indirectly associated with school adjust-
ment outcomes, mediated by the quality of the relationships that children
have with teachers and peers. Perhaps the behaviors that children exhibit
in the classroom influence the quality of the relationships that children
form with teachers and peers, and the quality of these relationships, in
turn, is predictive of young children's adjustment in school contexts.
Again, future investigations will elucidate these issues and, in fact, explo-
ration of this question is a central objective of our research program.
Other researchers have developed similar models to examine related is-
sues in older children. Wentzel (1993), for example, examined direct and
indirect connections between two types of behavior (prosocial and antiso-
cial), teacher preferences for students, and students' academic compe-
tence in sixth- and seventh-grades. The results indicate direct links be-
tween behavior and academic adjustment outcomes, but did not show
teacher preferences for students to be mediating a relation between behav-
ior and academic outcomes. Future studies that examine specific features
of the teacher-child relationship and use a broader definition of school ad-
justment, however, may yield different results.
In sum, in addition to describing the current state of the literature re-
garding the association between interpersonal features of the school envi-
ronment and children's adjustment in school contexts, we have also delin-
eated various avenues for future investigation, which will do much to
advance our understanding of the nature of these connections. The social
relationships children form with significant others in the school environ-
ment (i.e., peers and teachers) may serve important motivational functions
for young children as they form their initial opinions of and attitudes to-
wards school. If these relationships are characterized by acceptance, chil-
dren's motivation to be actively and appropriately engaged in the class-
room may be optimized. If, on the other hand, relationships with peers and
teachers are typified by conflict and rejection, children's motivation may
be suppressed or extinguished. Exploration of the aforementioned issues
will further illuminate these propositions.
Note
Preparation of this chapter was supported by National Institute of Mental
Health Grant MH-49223 to Gary Ladd. Correspondence concerning this
chapter should be addressed to Sondra Birch or Gary Ladd, 183 Children's
Research Center, 51 Gerty Drive, Champaign, Illinois 61820.
References
Ainsworth, M. D. S., Waters, E., & Wall, S. (1978). Patterns of attachment: A psy-
chological study of the strange situation. Hillsdale, NJ: Erlbaum.
Asarnow, J. R. (1983). Children with peer adjustment problems: Sequential and
nonsequential analysis of school behaviors. Journal of Consulting and Clinical
Asher, S. R., Singleton, L. C, Tinsley, B. R, & Hymel, S. (1979). A reliable so-
ciometric measure for preschool children. Developmental Psychology, 15,
443^44.
Berndt, T. J. (1989). Contributions of peer relationships to children's development.
In T. J. Berndt & G. W. Ladd (Eds.), Peer relationships in child development
(pp. 407-416). New York: Wiley.
Berndt, T. J., & Perry, T. B. (1986). Children's perceptions of friendships as sup-

portive relationships. Developmental Psychology, 22, 640—648.
Birch, S. H., & Ladd, G. W. (1994, April). The relative contributions of peer and
teacher-child relationships to children s early school adjustment. Paper pre-
sented at the 13th Biennial conference on Human Development, Pittsburgh, PA.
Birch, S. H., & Ladd, G. W. (1996a). Behavioral correlates of children s relation-
ships with classroom peers and teachers. Manuscript under review.
Birch, S. H., & Ladd, G. W. (1996b). The teacher-child relationship and children's
early school adjustment. Journal of School Psychology.
Bowlby, J. (1982). Attachment. New York: Harper Collins.
Bukowski, W. M., & Hoza, B. (1989). Popularity and friendship: Issues in theory,
measurement, and outcome. In T. J. Berndt & G. W. Ladd (Eds.), Peer relation-
ships in child development (pp. 15-45). New York: Wiley.
Bukowski, W. M., Hoza, B., & Newcomb, A. F. (1987). Friendship, popularity,
and the "self" during adolescence. Unpublished manuscript, University of
Maine, Orono.
Cantrell, V L., & Prinz, R. J. (1985). Multiple perspectives of rejected, neglec-
ted, and accepted children: Relation between sociometric status and behav-
ioral characteristics. Journal of Consulting and Clinical Psychology, 53, 884-
889.
Carlson, C. L., Lahey, B. B., & Neeper, R. (1984). Peer assessment of the social
behavior of accepted, rejected, and neglected children. Journal of Abnormal
Clark, K. E. (1994). Connectedness and autonomy in parent-child relationships:
Linkages to children s socioemotional orientation and peer relationships. Un-
published doctoral dissertation, University of Illinois, Urbana-Champaign.
Cassidy, J., & Asher, S. R. (1992). Loneliness and peer relations in young chil-
dren. Child Development, 63, 350-365.
Coie, J. D. (1990). Toward a theory of peer rejection. In S. R. Asher & J. D. Coie
(Eds.), Peer rejection in childhood (pp. 365-^01). New York: Cambridge Uni-
versity Press.
Coie, J. D., & Dodge, K. A. (1988). Multiple sources of data on social behavior
and social status in the school: A cross-age comparison. Child Development,
59, 815-829.
Coie, J. D., Dodge, K. A., & Coppotelli, H. (1982). Dimensions and types of so-
cial status: A cross-age perspective. Developmental Psychology, 18, 557-570.
Coie, J. D., Dodge, K. A., & Kupersmidt, J. B. (1990). Peer group behavior and
social status. In S. R. Asher & J. D. Coie (Eds.), Peer rejection in childhood
(pp. 17-59). New York: Cambridge University Press.
Coie, J. D., Finn, M., & Krehbiel, G. (1984, September). Controversial children:
Peer assessment evidence for status category distinctiveness. Paper presented at
annual meeting of the American Psychological Association, Toronto.
Coleman, C. C , & Ladd, G. W. (1993, April). How children who dislike school
feel about their classroom peer relationships. Paper presented at the annual
meetings of the American Educational Research Association, Atlanta, GA.
Sroufe (Eds.), Self processes in development: Minnesota Symposium on Child
Psychology (vol. 23, pp. 43-77). Hillsdale, NJ: Erlbaum.
Crick, N. R., & Ladd, G. W. (1993). Children's perceptions of their peer experi-
ences: Attributions, loneliness, social anxiety, and social avoidance. Develop-
Dodge, K. A. (1983). Behavioral antecedents of peer social status. Child Develop-
ment, 54, 1386-1399.
Dodge, K. A., Coie, J. D., & Brakke, N. P. (1982). Behavior patterns of socially re-
jected and neglected preadolescents: The roles of social approach and aggres-
sion. Journal of Abnormal Child Psychology, 10, 389-410.
Feldhusen, J. E, Thurston, J. R., & Benning, J. J. (1970). Longitudinal analyses of
classroom behavior and school achievement. Journal of Experimental Educa-
tion, 55,4-10.
Furman, W., & Buhrmester, D. (1995). Children's perceptions of the personal rela-
tionships in their social networks. Developmental Psychology, 21, 1016-1021.
Gershman, E. S., & Hayes, D. S. (1983). Differential stability of reciprocal friend-
ships and unilateral relationships among preschool children. Merrill-Palmer
Quarterly, 29, 169-177.
Goldman, J. A., Corsini, D. A., & deUrioste, R. (1980). Implications of positive
and negative sociometric status for assessing the social competence of young
children. Journal of Applied Developmental Psychology, 1, 209-220.
Green, K. D., Forehand, R., Beck, S., & Vosk, B. (1980). An assessment of the re-
lationship among measures of children's social competence and children's aca-
demic achievement. Child Development, 51, 1149-1156.
Hart, C. H., Ladd, G. W., & Burleson, B. R. (1990). Children's expectations of the
outcomes of social strategies: Relations with sociometric status and maternal
disciplinary styles. Child Development, 61, 127-137.
Hartup, W. H. (1989). Social relationships and their developmental significance.
A merican Psychologist, 44, 120—126.
Hinde, R. A., Titmus, G., Easton, D., & Tamplin, A. (1985). Incidence of "friend-
ship" and behavior toward strong associates versus nonassociates in preschool-
ers. Child Development, 56, 234-245.
Hoover, J., & Hazier, R. J. (1991). Bullies and victims. Elementary School Guid-
ance and Counseling, 25, 212—219.
Howes, C. (1983). Patterns of friendship. Child Development, 54, 1041-1053.
Howes, C. (1988). Peer interaction of young children. Monographs of the Society
for Research in Child Development, 55 (1, Serial No. 217).
Howes, C , & Hamilton, C. E. (1992). Children's relationships with child care
teachers: Stability and concordance with parental attachments. Child Develop-

ment, 63, 867-878.
Howes, C, & Hamilton, C. E. (1993). The changing experience of child care:
Changes in teachers and in teacher-child relationships and children's social
competence with peers. Early Childhood Research Quarterly, 8, 15-32.
Howes, C, & Matheson, C. C. (1992). Contextual constraints on the concordance
of mother-child and teacher-child relationships. New Directions for Child De-
velopment, 57, 25-40.
Howes, C, Matheson, C. C, & Hamilton, C. E. (1994). Maternal, teacher, and
child care history correlates of children's relationships with peers. Child Devel-
opment, 65, 264-273.
Howes, C, Phillips, D. A., & Whitebrook, M. (1992). Thresholds of quality in
child care centers and children's social and emotional development. Child De-
velopment, 63, 449^460.
Kochenderfer, B. J., & Ladd, G. W. (in press). Peer victimization: Manifestations
and relations to school adjustment in kindergarten. Journal of School Psycholo-
gy-
La Greca, A. M. (1981). Peer acceptance: The correspondence between children's
sociometric scores and teachers' ratings of peer interactions. Journal of Abnor-
mal Child Psychology 9, 167-178.
Ladd, G. W. (1983). Social networks of popular, average, and rejected children in
school settings. Merrill-Palmer Quarterly, 29, 283-307.
Ladd, G. W. (1989). Children's social competence and social supports: Precursors
of early school adjustment? In B. H. Schneider, G. Attili, J. Nadel, & R. P.
Weissberg (Eds.), Social competence in developmental perspective (pp.
277-291). Amsterdam: Kluwer Academic Publishers.
Ladd, G. W. (1990). Having friends, keeping friends, making friends, and being
liked by peers in the classroom: Predictors of children's early school adjust-
ment? Child Development, 61, 1081-1100.
Ladd, G. W. (1996). Shifting ecologies during the 5-7 year period: Predicting chil-
dren's adjustment during the transition to grade school. In A. Sameroff & M.
Haith (Eds.), The Five to Seven Year Shift (pp. 363-386). Chicago: University
of Chicago Press.
Ladd, G. W., & Emerson, E. S. (1984). Shared knowledge in children's friend-
ships. Developmental Psychology, 20, 932-940.
Ladd, G. W., & Goiter, B. S. (1988). Parents' management of preschoolers' peer
relations: Is it related to children's social competence? Developmental Psychol-
ogy 24, 109-117.
Ladd, G. W., & Hart, C. H. (1992). Creating informal play opportunities: Are par-
ents' and preschoolers' initiations related to children's competence with peers?
Developmental Psychology 28, 1179-1187.
Ladd, G. W., & Kochenderfer, B. J. (1996). Linkages between friendship and ad-
justment during early school transitions. In W. M. Bukowski, A. F. Newcomb,

& W. W. Hartup (Eds.), The company they keep: Friendship in childhood and
adolescence (pp. 322-345). New York: Cambridge University Press.
Ladd, G. W., Kochenderfer, B. J., & Coleman, C. C. (in press). Friendship quality
as a predictor of young children's early school adjustment. Child Development.
Ladd, G. W., & Price, J. M. (1987). Predicting children's social and school adjust-
ment, 58, 1168-1189.
Lazar, I., & Darlington, R. (1982). Lasting effects of early education: A report
from the Consortium for Longitudinal Studies. Monographs of the Society for
Research in Child Development, 47, (2-3), Serial Number 195.
Lynch, M, & Cicchetti, D. (1992). Maltreated children's reports of relatedness to
their teachers. New Directions for Child Development, 57, 81-108.
Marshall, H. R., & McCandless, B. R. (1957). Relationships between dependence
on adults and social acceptance by peers. Child Development, 28, 413^419.
Moore, S., & Updegraff, R. (1964). Sociometric status of preschool children relat-
ed to age, sex, nurturance-giving, and dependency. Child Development, 35,
519-524.
Parker, J. G., & Asher, S. R. (1993a). Beyond peer group acceptance: Friendship
and friendship quality as distinct dimensions of peer adjustment. In W. H. Jones
& D. Perlman (Eds.), Advances in personal relationships (vol. 4, pp. 261—294).
London: Kinglsley Press.
Parker, J. G., & Asher, S. R. (1993b). Friendship and friendship quality in middle
childhood: Links with peer group acceptance and feelings of loneliness and so-
cial dissatisfaction. Developmental Psychology, 29, 611-621.
Pianta, R. C, & Steinberg, M. (1992). Teacher-child relationships and the process
of adjusting to school. New Directions for Child Development, 57, 61-80.
Pianta, R. C, Steinberg, M., & Rollins, K. (1995). The first two years of school:
Teacher-child relationships and deflections in children's school adjustment. De-
velopment and Psychopatho logy, 7, 295-312.
Profilet, S. M., & Ladd, G. W. (1994). Do mothers' perceptions and concerns
about preschoolers' peer competence predict their peer management practices?
Social Development, 3, 205-221.
Rabiner, D. L., & Coie, J. D. (1989). Effect of expectancy inductions on rejected
children's acceptance by unfamiliar peers. Developmental Psychology, 25,
450-^57.
Rubin, K. H. (1982). Social and social-cognitive developmental characteristics of
young isolate, normal, and sociable children. In K. H. Rubin & H. S. Ross
(Eds.), Peer relationships and social skills in childhood (pp. 353-374). New
York: Springer-Verlag.
Ryan, R. M., & Powelson, C. L. (1991). Autonomy and relatedness as fundamen-

tal to motivation and education. Journal ofExperimental Education, 60, 49-66.
Safran, S. P., & Safran, J. S. (1985). Classroom context and teachers' perceptions
of problem behaviors. Jorunal of Educational Psychology, 77, 20-28.
Skinner, E. A., & Belmont, M. J. (1993). Motivation in the classroom: Reciprocal
effects of teacher behavior and student engagement across the school year.
Vandell, D. L., & Mueller, E. C. (1980). Peer play and friendship during the first
two years. In H. C. Foot, A. J. Chapman, & J. R. Smith (Eds.), Friendship and
social relations in children (pp. 181-208). New York: Wiley.
Weiss, R. (1974). The provisions of social relationships. In Z. Rubin (Ed.), Doing
unto others (pp. 17-26). Englewood Cliffs, NJ: Prentice-Hall.
Wentzel, K. R. (1991). Social competence at school: Relation between social re-
1-24.
Wentzel, K. R. (1993). Does being good make the grade? Social behavior and aca-
demic competence in middle school. Journal of Educational Psychology, 85,
357-364.
Wentzel, K. R. (1994). Relations of social goal pursuit to social acceptance, class-
room behavior, and perceived social support. Journal of Educational Psycholo-
gy, 86, 173-182.
10 Social goals and social relationships
as motivators of school adjustment
Kathryn R. Wentzel
Why are some children eager to learn and master new intellectual chal-
lenges while others devalue and disengage from academic activities? Why
do some children become energized by intellectual challenges while oth-
ers shrug their shoulders in the face of academic failure? Models of
achievement motivation most often attribute these distinct motivational
orientations to cognitive processes believed to regulate behavior. These
cognitive processes have been studied extensively with regard to attribu-
tions for success and failure (Weiner, 1985), mastery and performance
goal orientations (Nicholls, 1984), beliefs about the nature of intelligence
(Dweck & Leggett, 1988), and beliefs about values and ability (e.g., Ban-
dura, 1986; Eccles, Adler, Futterman, Goff, Kaczala, Meece, & Midgley,
1983). How and why these belief systems develop have not been studied
as frequently, although classroom interventions that provide mastery-ori-
ented reward structures and activities that are meaningful and interesting
to students appear to alleviate some motivational deficits.
In general, the work that has grown out of these models assumes that
intellectual competence is the primary goal that children try to achieve at
school, and that children's reasons for why they try to achieve academical-
ly are the key predictors of their learning-related behavior. There is a
growing body of evidence, however, that a consideration of the social
worlds of children should not be excluded from models of classroom mo-
tivation if we are to understand children's successes and failures at school.
In fact, throughout the history of American education, social competen-
cies have been among the most critical objectives that children are expect-
ed to achieve at school (see Wentzel, 1991b). Moreover, social goals and
behavior are strong and consistent predictors of academic outcomes, and
the social climate of the classroom appears to be a powerful motivator of
academic as well as socially appropriate classroom behavior.
The purpose of this chapter, therefore, is to highlight children's proso-
226
Social and academic motivation 227
cial and socially responsible classroom behavior as one aspect of social

competence that is also a critical component of school adjustment. In ad-
dition, I will present evidence to suggest that students who pursue multi-
ple goals reflecting social as well as academic objectives are those who
are most successful at school. Finally, socialization processes that con-
tribute to social goal pursuit will be discussed.
Prosocial and responsible behavior as

an aspect of school adjustment
The ability and willingness to help, share, cooperate, and comply with
rules and role expectations represent aspects of social competence that
serve to maintain social cohesion and the integrity of social groups over
time. Displays of prosocial and responsible behavior are critical for the
development of positive social relationships within the peer group (Hart-
up, 1983) as well as the family system (Maccoby & Martin, 1983). Of par-
ticular importance for understanding children's school success is that re-
search has documented links between prosocial and responsible behavior
and children's intellectual functioning. For instance, studies have linked
positive intellectual outcomes in the elementary years with tendencies to
be prosocial and empathic, prosocial interactions with peers, appropriate
classroom conduct, and compliance (see Wentzel, 1991b). In a meta-
analysis of factors related to early learning problems, social and emotion-
al variables predicted achievement as well as or better than intellectual
ability, sensory deficits, or neurological factors (Horn & Packard, 1985).
Prosocial and responsible behavior has been linked to academic ac-
complishments in the middle school years as well. In a series of studies on
young adolescents, self-restraint has been related consistently and posi-
tively to children's academic accomplishments (Feldman & Wentzel,
1990; Wentzel, 1994a; Wentzel & Feldman, 1993; Wentzel, Weinberger,
Ford, & Feldman, 1990). Self-restraint represents one aspect of socially
responsible behavior in that it refers to self-regulation in the interest of
achieving a balance between personal needs and those of others. Prosocial
and responsible classroom behavior also has been related directly to class-
room grades and test scores (Wentzel, 1991a, 1993b), even when the po-
tentially confounding effects of academic behavior, teachers' preferences
for students, IQ, family structure, sex, ethnicity, and days absent from
school were taken into account (Wentzel, 1993b).
Longitudinal studies linking prosocial and socially responsible behav-
ior to academic achievement have been less frequent, but have yielded the
same general findings in that adaptive classroom behavior in elementary

school predicts grades and test scores in high school, over and above early
achievement or IQ (Feldhusen, Thurston, & Benning, 1970; Lambert,
1972; Safer, 1986). Findings from longitudinal studies also link social re-
sponsibility with educational attainment. Based on a comprehensive re-
view of both follow-up and follow-back studies, Parker and Asher (1987)
conclude that antisocial and aggressive (i.e., socially irresponsible) behav-
ior in the early grades is a strong predictor of dropping out in high school.
Finally, intervention studies suggest that teaching children appropriate so-
cial responses to instruction such as attending, following instructions, and
volunteering answers can lead to significant and stable gains in academic
achievement (Cobb & Hopps, 1973; Coie & Krehbiel, 1984; Hopps &
Cobb, 1974).
Summary and implications

Displays of prosocial and socially responsible behavior (e.g., tendencies
to be cooperative, helpful, compliant, and nonaggressive) have been relat-
ed consistently and positively to children's academic accomplishments.
Several explanations for these associations are possible. First, social com-
petence might set the stage for academic competencies to develop. Sec-
ond, academic competencies might lead to displays of socially appropriate
behavior. Third, social and academic competencies might influence each
other in reciprocal fashion. Finally, social and academic competence
might not be causally related but rather, an additional variable or set of
variables might account for the significant correlations between social and
academic outcomes.
There is scant evidence to support any one of these explanations. The
longitudinal studies provide initial evidence in support of the first model.
In addition, intervention studies indicate that the promotion of socially re-
sponsible behavior often results in higher levels of academic performance
rather than the reverse (Cobb & Hopps, 1973; Coie & Krehbiel, 1984;
Hops & Cobb, 1974). Research on the underlying processes that might
contribute to social as well as academic outcomes also has not been exten-
sive. However, from a motivational perspective, those processes that ap-
pear to motivate children to behave in prosocial and responsible ways also
appear to motivate academic performance. In the following section, social
goal pursuit will be discussed as a potentially critical motivational process
that links children's social adjustment to their academic performance at
school.
Social goal pursuit and children's adjustment at school

Personal goals have been of central importance in explanations of motiva-
tional orientations and patterns of behavior (see Pervin, 1982; Ford, this
volume). Goals have been described with respect to their content (Ford &
Nichols, 1991; Wentzel, 1993a, 1994b), orientation (Dweck & Leggett,
1988; Nicholls, 1984), levels of challenge, proximity, and specificity
(Bandura, 1986), and their relations to each other (Wentzel, 1991c). Cen-
tral to all of these descriptions is the notion that people do set goals for
themselves and, as cognitive representations of future events, these goals
can be powerful motivators of behavior.
The notion that goals organize and regulate behavior also has been
central to research linking achievement motivation to academic perfor-
mance and learning-related outcomes. Students' goal orientations most
often associated with academic outcomes are described as concerns with
either task-intrinsic or extrinsic outcomes. Task-intrinsic outcomes such
as task-involved (Nicholls, 1984) or learning (Dweck & Leggett, 1988)
goals represent outcomes reflecting the actual process of learning. Such
goals can take the form of increased knowledge or task mastery. Task-ex-
trinsic goals such as ego-involved (Nicholls, 1984) or performance
(Dweck & Leggett, 1988) goals are those derived from social expectations
or values associated with the consequences of task performance. In this
case, desired outcomes can take the form of praise or tangible rewards.
For the most part, research on achievement goal orientations has gener-
ated discussions concerning the educational benefits of task-involved and
learning goals relative to ego-involved and performance goals. Several au-
thors, however, have focused on the broader social context of the class-
room, suggesting that school adjustment requires the pursuit of multiple
and often complementary goals, both social and academic. Early work in
this area identified the need for approval as an important social motivatior
of classroom learning (e.g., Maehr, 1983; Veroff, 1969). More recent work
has recognized a broader range of social concerns as important aspects of
school-related motivation. For instance, Kozeki (1985; Kozeki & En-
twisle, 1984) proposes three dimensions of school motivation: an affective
domain represented by motives to maintain relationships with parents,
teachers, and peers; a moral domain represented by motives to establish
trust, compliance, and social responsibility; and a cognitive domain repre-
sented by motives for independence, competence, and interest.
Most recently, Connell and his colleagues (Connell & Wellborn 1991;
Deci & Ryan, 1991) have developed a model of student motivation based
on three basic needs: competence, autonomy, and relatedness. With re-

spect to social aspects of motivation, Connell and Wellborn (1991) sug-
gest that a sense of social relatedness contributes to the adoption of goals
promoted by social groups or institutions, whereas a lack of relatedness or
disaffection can lead to a rejection of such goals. In other words, the pur-
suit of social goals that promote group cohesion and positive interperson-
al interactions (such as to be prosocial and responsible), depends in large
part on feeling like one is an integral part of the social group.
In the following sections, the literature on children's pursuit of social
goals at school will be reviewed. Then, relations between social goal pur-
suit and academic accomplishments will be discussed.
Children s pursuit of social goals at school

In my own work, I have taken the perspective that as in other socialization
contexts, children are required to pursue goals to conform to rules for so-
cial conduct and to behave in socially appropriate ways when they are at
school. Moreover, I view the classroom as a social context that affords stu-
dents the opportunity to pursue other social goals related to social interac-
tion and relationship development. Empirical work on social goal pursuit
as an integral part of classroom life has not been extensive. However, Ford
(1982) found that the pursuit of goals to establish and maintain social rela-
tionships is an important school-related activity for adolescents. In an
ethnographic study of 9th graders, Allen (1986) also found that the pursuit
of socializing goals is an integral part of classroom life. Similarly, in a
study of 9th through 12th graders, I found that adolescents reported trying
to achieve social goals to make friends, have fun, and be dependable and
responsible more often than learning-related goals (Wentzel, 1989).
In recent work on students' pursuit of prosocial and responsibility
goals, I have found that middle school students also pursue prosocial and
responsibility goals with greater frequency than either learning-related
goals or goals to socialize with peers (Wentzel, 1991c, 1992). In this re-
search, social goal pursuit was operationalized as students' self-reported
efforts to help, share, and cooperate with teachers and peers (prosocial
goals) and to follow rules and keep interpersonal promises and commit-
ments (social responsibility goals). Pursuit of these social goals has also
been related to social acceptance by classmates and teachers (Wentzel,
1991c, 1994b). Social acceptance among classmates was measured using
sociometric techniques designed to identify popular, rejected, neglected,
controversial, and average status children (see Wentzel, 1991a). To assess
teachers' acceptance of students, teachers were asked to rate how much

they would like to have each of their students in their class again next year.
When compared with average status children, popular children (those
with many friends and who are disliked by few of their classmates) report-
ed trying to achieve prosocial goals significantly more often, neglected
children (those who have few friends but are not disliked) reported trying
to achieve both prosocial and compliance goals significantly more often,
and controversial children (those with many freinds but who are also dis-
liked by many of their peers) reported trying to achieve compliance goals
significantly less often. In the case of teacher acceptance, students report-
ing high levels of effort to achieve social responsibility goals were pre-
ferred by teachers significantly more than other students (see Wentzel,
1991a).
Social goal pursuit and academic outcomes

Findings linking social goal pursuit to academically-relevant behavior
have been inconsistent. Early work in this area suggests that the pursuit of
goals to be compliant and responsible is associated with outcomes often
antagonistic to learning (Kozeki & Entwistle, 1984; Kozeki, 1985). In
contrast, other findings suggest positive relations between a need for ap-
proval and academic outcomes depending on the age of students (Cran-
dall, 1966; Veroff, 1969), and whether these goals were pursued in con-
junction with or in opposition to more academically-relevant goals
(Nakamura & Finck, 1980).
In my own research, the pursuit of goals to behave in prosocial and so-
cially responsible ways has been related consistently and positively to aca-
demic motivation as well as performance. In a study of adolescents from a
suburban high school (Wentzel, 1989), students' self-reported efforts to
achieve six classroom goals were related significantly and positively to
student grades. These goals represented social as well as cognitive out-
comes: to be a successful student, be dependable and responsible, learn
new things, understand things, do your best, and get things done on time.
Self-reported efforts to have fun were related negatively to students'
grades.
In addition, relations between academic performance and student re-
ports of effort could be explained more precisely by the sets of goals that
students tried to achieve at school. In particular, 84% of the highest
achieving students reported always trying to be a successful student, to be
dependable and responsible, and to get things done on time; only 13% of
the lowest achieving students reported always trying to achieve these three
goals. Moreover, although the highest achieving students reported fre-
quent pursuit of academic goals (i.e., to learn new things, to understand
things), less frequent pursuit of these goals did not distinguish the lowest
achieving from average achieving students. Rather, an unwillingness to
try to conform to the social and normative standards of the classroom
uniquely characterized the lowest achieving students.
A second study of 6th and 7th grade middle school students and their
teachers was conducted in a predominantly working class, midwestern
community (Wentzel, 1993a). In this research, two academic goals reflect-
ing efforts to master new and challenging tasks and to earn positive evalu-
ations, and two social goals reflecting efforts to be prosocial and to be so-
cially responsible were investigated. As in the first study, I found that goal
pursuit differed as a function of students' levels of academic achievement.
Findings also indicated a significant relation between combinations of
goal pursuit and academic achievement: 59% of the high-achieving stu-
dents reported frequent efforts to achieve both social and academic goals
whereas only 38% of the average achievers and 34% of the low-achieving
students reported similar levels of effort to achieve these goals.
Summary and implications

There is growing evidence to suggest that students' pursuit of prosocial
and social responsibility goals is related to their levels of social accep-
tance by peers and teachers. In addition, results linking the pursuit of
prosocial and social responsibility goals to the pursuit of academic goals
as well as to actual levels of academic performance suggest that social
goal pursuit might be a necessary (albeit insufficient) requirement for
achievement in both social and academic domains of functioning. Given
the seemingly critical role of prosocial and socially responsible goal pur-
suit and behavior for multiple aspects of classroom adjustment, it there-
fore becomes important to understand the socialization processes that
might contribute to the development of these motivational orientations
and to their expression in classroom settings.
Research on classroom factors that directly promote the pursuit of
these social goals is rare. However, the literature on family socialization
suggests that behaving in prosocial and responsible ways might represent
an important social competency that links the quality of family function-
ing and parent-child relationships to children's school adjustment. In the
following section, I will review this literature, adopting a perspective that
children's tendencies to behave in socially appropriate ways at school can

be understood by examining socialization experiences at school that paral-
lel those that occur within the context of family relationships.
Socialization processes contributing to

social goal pursuit at school
In the developmental literature, models of self and social skills develop-
ment have posited that personal goals play a key role in the organization
and direction of social information processing and behavioral regulation
(Dodge, 1986; Dweck & Leggett, 1988; Ford, 1992). Models of socializa-
tion stress the process of adopting and internalizing societal and parental
goals as a critical part of children's socialization into adult roles (see
Grusec & Goodnow, 1994; Maccoby & Martin, 1983). The ability to pur-
sue multiple goals, especially the ability to resolve conflicts between per-
sonal goals and those of others, is considered to be an integral part of this
process. In the following sections, socialization processes that might ex-
plain children's pursuit of social goals at school will be discussed.
Models of socialization within the family

Perhaps most relevant for understanding why children are motivated to be-
have in socially appropriate ways are models of socialization that focus on
processes that promote children's general compliance and internalization
of rules and norms for behavior. In support of a motivational interpreta-
tion, Maccoby (1992) suggests that "socialization does not merely involve
the inculcation of socially acceptable habits, but also of motives, so that
children willingly adopt and enact appropriate behavior patterns without
constant surveillance . . ." (p. 172).
Research based on family socialization models typically has examined
the quality of parent-child interactions and relationships, or specific
parental disciplinary styles, in relation to a broad range of children's so-
cial, emotional, and cognitive outcomes (see Maccoby & Martin, 1983).
Baumrind, for instance, found that four dimensions of parent-child inter-
actions could predict reliably children's social adjustment (Baumrind,
1971; 1991). Parental control reflects consistent enforcement of rules,
provision of structure to children's activities, and persistence in gaining
child compliance; maturity demands reflect expectations to perform up to
one's potential, and demands for self-reliance and self-control; clarity of
communication reflects the extent to which parents solicit children's opin-
ions and feelings, and use reasoning to obtain compliance; and nurturance
reflects parental expressions of warmth and approval as well as conscien-
tious protection of children's physical and emotional well-being. Other
family socialization models are compatible with this approach (see
Grusec & Goodnow, 1994).
Research based on these models suggests that a child is more likely to
behave in prosocial and responsible ways if her parents are warm and nur-
turing, responsive to her wishes and point of view, and minimize external
pressures to comply by using inductive reasoning rather than power as-
sertive methods of discipline (Baumrind, 1971, 1991; Maccoby & Martin,
1983). It is beyond the scope of this chapter to discuss the precise mecha-
nisms by which these practices lead to the internalization of social goals.
In general, however, it is believed that these aspects of parent-child inter-
actions contribute to the pursuit of socially valued goals because they (1)
provide the structure necessary for learning appropriate forms of social
behavior; (2) communicate expectations that a child's behavior will reflect
his best intentions and abilities; (3) provide opportunities for autonomous
decision making and the development of feelings of self-determination;
and (4) support the development of positive feelings of self-worth (Grusec
& Goodnow, 1994).
Interestingly, this work has not identified mechanisms that can explain
the generalization and transfer of socially competent behavior from the
family setting to that of the classroom. On the one hand, socialization
models would suggest that goals internalized as a result of family social-
ization processes motivate children to behave in prosocial, responsible
ways in nonfamilial contexts (see Wentzel, 1994a). On the other hand,
children's goals and behavior are not always consistent across settings.
Therefore, alternative or complementary models of classroom socializa-
tion need to be developed that can explain classroom-specific social com-
petence.
One way to address the issue of transfer and generalization is to identi-
fy social processes in the classroom that correspond to those known to
promote positive social behavior and the internalization of socially inte-
grative goals within the family. Recent research indicates that children's
relationships with peers and teachers might provide them with valuable
resources similar to those provided by relationships with parents. In the
following section, this literature will be reviewed and parallels between
family and school socialization processes will be drawn. In particular, the
degree to which teachers and peers provide stable and consistent structure
and guidance, opportunities for democratic decision making, high stan-

dards for performance, and nurturance will be discussed.
Socialization processes in the classroom

Research on classroom processes that directly correspond to those known
to promote prosocial and socially responsible behavior within the context
of parent-child relationships has not been extensive. However, the history
of American education suggests that the development of socially integra-
tive skills has long been a primary function of the schooling process (e.g.,
Dreeben, 1968; Jackson, 1968; Wentzel, 1991b). At the policy level, so-
cially responsible and prosocial behavior in the form of moral character,
conformity to social rules and norms, cooperation, and positive styles of
social interaction has been a stated educational objective for American
schools. In fact, character development and social responsibility in gener-
al have been explicit objectives for public schools in almost every educa-
tional policy statement since 1848, being promoted with the same fre-
quency as the developement of academic skills (see Wentzel, 1991b).
The suggestion that the development of prosocial and socially respon-
sible behavior is an important educational goal is also supported by litera-
ture indicating that teachers as well as students provide socialization expe-
riences that promote displays of socially competent behavior. For
example, with respect to Baumrind's parenting dimensions, instances of
control, maturity demands, and democratic communication styles have
been observed on the part of teachers and students. Indeed, teachers ap-
pear to have common social rules and norms that they expect students to
follow (e.g., Hargreaves, Hester, & Mellor, 1975). Blumenfeld and her
colleagues have documented teacher communications to students con-
cerning why they ought to behave in socially appropriate ways and what
will happen if they do not (Blumenfeld, Hamilton, Bossert, Wessels, &
Meece, 1983). Shultz and Florio (1979) describe how teachers indicate to
students when they need to pay attention as a function of which contexts
they are in. Although rarely linked to academic performance, teacher pro-
vision of structure and guidance has been related positively to elementary
school-aged students' engagement in classroom activities (Skinner & Bel-
mont, 1993).
For the most part, classroom rules and teacher communications are de-
signed to establish classroom order and the hierarchical nature of teacher-
student relationships. However, teachers establish norms for peer relation-
ships that also focus on considerate, cooperative, and morally responsible

forms of behavior (Hargreaves et al., 1975; Sieber, 1979). Students them-
selves have been observed to contribute to social order in the classroom.
This occurs most often when they are held accountable for the behavior of
each other or when teachers use peer group leaders to monitor the class
when they must leave their classroom (Seiber, 1979). Students also have
been observed to monitor each other by ignoring irrelevant or inappropri-
ate behavior and responses during group instruction, and by private sanc-
tioning of inappropriate conduct (Sieber, 1979).
Children also appear to provide each other with instrumental help and
guidance that directly facilitates adherence to teachers' expectations for
behavior and academic performance. For instance, students frequently
clarify and interpret their teacher's instructions and provide mutual assis-
tance (Sieber, 1979). Classmates also provide each other with information
by modeling both academic and social competencies and establishing nor-
mative standards for performance (Schunk & Zimmerman, this volume).
Finally, teachers and peers can also be valuable sources of nurturance.
Studies of perceived social support from peers suggest that young adoles-
cents are primarily sources of companionship for each other (Burmester &
Furman, 1987; Reid, Landesman, Treder, & Jaccard, 1989; Youniss &
Smollar, 1989). Within the context of these relationships, however, peers
tend to provide each other with nurturance and emotional support by serv-
ing as buffers against stress in unfamiliar surroundings and bolstering per-
ceptions of self-worth and self-esteem. With respect to academic out-
comes, evidence suggests that positive relationships with peers can
provide emotional security and incentives to achieve (e.g., Ladd & Price,
1987), whereas the loss of a familiar peer group can have negative effects
on self-esteem and general interest in school (Miller, 1983). Also, per-
ceived isolation from peers as well as perceived lack of control in obtain-
ing social support at school have been related to low levels of achievement
(Epperson, 1963).
Although research on social support from teachers has been less fre-
quent (see Birch & Ladd and Harter, this volume, for exceptions), young
adolescents report that teachers can be valuable sources of help and guid-
ance outside the home (Burmester & Furman, 1987). However, young
adolescents report declines in the quality of teacher-student relationships
after the transition to middle school; these declines correspond to declines
in academic motivation and achievement (Feldlaufer, Midgley, & Eccles,
1988; Midgley, Feldlaufer & Eccles, 1989).
Experiences with peers and teachers as motivators

of social goal pursuit
The literature reviewed in the previous section describes ways in which
teachers and peers provide students with socialization experiences con-
ducive to the development of socially competent classroom behavior. Few
studies have documented relations between these socialization experi-
ences and student motivation (see Berndt & Keefe and Birch & Ladd, this
volume, for exceptions). As suggested by models of family socialization,
however, one reason why social interactions with teachers and peers might
be related to aspects of school adjustment is that they promote positive
motivational orientations toward prosocial behavior and respect for class-
room rules and norms.
My recent work has begun to explore ways in which students' pursuit
of goals to be prosocial and responsible is related to interpersonal re-
sources from teachers and peers (Wentzel, 1994b). In this research, inter-
personal support is reflected in perceptions that peers and teachers care
about the student as a person and as a learner. Results indicate that middle
school students' pursuit of goals to help and cooperate (prosocial goals)
and to follow classroom rules and norms (social responsibility goals) is
related to perceived social support from teachers as well as peers. These
findings were robust even when taking into account actual levels of social
acceptance by teachers and peers, and students' sex, ethnicity, and family
structure. In a second study of middle school students, perceived support
from teachers and peers was related negatively to students' emotional dis-
tress and positively to interest in school (Wentzel, 1996).
These findings support the notion that interpersonal support from
teachers and peers (i.e., perceiving that teachers and peers care about
them) are related positively to social and motivational aspects of school
adjustment (see also Goodenow, 1993; Midgley et al., 1989; Wentzel &
Asher, 1995). In an effort to relate these results to a family socialization
perspective, I conducted a third study that examined students' characteri-
zations of "caring." I asked students to list three things that teachers do to
show that they care about them and three things that teachers do to show
that they don't care. Students then repeated this process for peers.
For teachers, student responses were coded along dimensions reflect-
ing those typically studied in the parenting literature. Responses reflecting
three of the four dimensions emerged from the data: democratic interac-
tion styles as demonstrated by respect, recognition of social and academic
Table 10.1. Students' descriptions of caring and uncaring teachers

Percentage of
Characteristics of caring teachers total responses
Democratic interaction styles
Communication 25
Respect 4
Recognition of individual differences
Personal 15
Academic 13
Expectations for achievement 30
Positive encouragement and feedback 11
Other (e.g., generic helping, being nice) 6
Characteristics of uncaring teachers
Lack of democratic interaction styles
Lack of communication 46
Lack of respect 9
Lack of recognition of individual differences
Personal 8
Academic 7
Low expectations for achievement 18
Criticism and negative feedback 10
Other (e.g., negative personality traits) 2
individual differences, and solicitation of student input and opinions; high

expectations for achievement; and positive encouragement and feedback
(see Table 10.1). These categories of responses correspond to Baumrind's
dimensions of clarity of communication, maturity demands, and nurtur-
ance, respectively.
In contrast to adult-child relationships that are predominantly hierar-
chical, the nature of peer relationships is relatively egalitarian. Therefore,
dimensions of support from peers differ somewhat from those of parents
or other adults. Consequently, characterizations of "caring" and "uncar-
ing" peers were coded along four dimensions derived from work by
Burmester & Furman (1987): companionship, help, reliable alliance, and
emotional support (see Table 10.2). However, provision of help and a sta-
ble and reliable source of guidance correspond roughly to parental control
in that both represent a source of consistent and reliable support. Emo-
tional support corresponds to nurturance in that it represents concern with
promoting and maintaining positive levels of emotional well-being.
The findings shown in Tables 10.1 and 10.2 are relevant for under-
standing ways in which socialization experiences in the classroom parallel
Table 10.2. Students' descriptions of caring and uncaring peers

Percentage of
Characteristics of caring peers total responses
Companionship 32
Academic help 24
Reliable allies 15
Emotional support (positive regard & concern) 29
Characteristics of uncaring peers

Lack of companionship 17
Lack of academic help 9
Unreliable allies 13
Emotionally unsupportive
Lack of concern 8
Verbal and physical aggression 53
those of the family. Specifically, these results indicate that students char-
acterize positive relationships with peers and teachers in ways that are
similar to positive parent-child relationships. Of interest, however, is that
students tend to perceive teachers and peers as sources of different types
of experiences and support. Students seem to look to teachers primarily
for opportunities for self-expression and autonomous decision making
(democractic interaction styles), and for standards for performance. In
contrast, students tend to look to peers primarily for companionship and
emotional support.
Of final interest is that preliminary findings indicate that these dimen-
sions of social support from teachers and peers are related to social and
motivational outcomes. Specifically, students who define uncaring teach-
ers as failing to provide nurturance and autonomy support are not highly
motivated to behave in prosocial and socially responsible ways, whereas
those who define caring peers as emotionally supportive are those who
tend to be motivated to be prosocial and socially responsible.
Conclusions and future directions

The literature reviewed in this chapter suggests that social competence in
the form of prosocial and socially responsible behavior is a critical aspect
of school adjustment that is valued by teachers as well as students. Proso-
cial and responsible behavior also has been linked consistently with acad-
emic accomplishments. Second, students who pursue multiple goals re-
fleeting social as well as academic objectives are those who are most suc-
cessful at school. Moreover, pursuit of goals to behave in prosocial and re-
sponsible ways appear to be an underlying variable that links social ad-
justment with positive academic outcomes. Finally, students' pursuit of
these goals appears to be motivated, in part, by the quality of support de-
rived from relationships with teachers and peers.
In addition, I have proposed that access to interpersonal resources and
social support promotes prosocial and socially responsible behavior in
both family and school settings. This proposition, however, reflects sever-
al important assumptions. First, it assumes that all children value proso-
cial and socially responsible behavior and that all that is necessary to elic-
it pursuit of goals to behave in this manner is the existence of certain
interpersonal resources. Related to this notion is the assumption that pro-
visions of specific types of support by teachers and peers will provide stu-
dents with continuity between home and school. Finally, it is assumed that
once a child adopts goals to be prosocial and socially responsible, she will
be able to achieve those goals. Each of these assumptions will be dis-
cussed in turn.
Where do goals come from?

A central theme of this chapter has been that children's willingness and
desire to pursue goals to be cooperative and compliant is an important as-
pect of social motivation that contributes to the stability and maintenance
of their respective social groups. Little is known, however, about how chil-
dren come to value these particular goals in the first place. As suggested
in the socialization literature, children will adopt those goals that are val-
ued by individuals with whom they have warm and supportive relation-
ships. Therefore, an examination of the content of family goals and how
children come to adopt them as their own seems to be especially impor-
tant for understanding children's reactions to and adoption of social rules
and norms in other social settings such as the classroom.
Research on the content of family goals might be especially relevant
for understanding the achievement of children from minority cultures who
are expected to comply with rules and norms for classroom behavior that
might be inconsistent with those espoused by their families and communi-
ty (see, e.g., Ogbu, 1985). Alternatively, research on goals promoted by
supportive peer groups might also help to explain the motivational orien-
tations of children who do not experience warm and nurturant relation-
ships with their parents.
Developmental issues also need to be considered. For instance, the

transition from elementary to middle school is often marked by height-
ened levels of mistrust between teachers and students, student perceptions
that teachers no longer care about them, and a decrease in opportunities
for students to establish meaningful relationships with teachers and peers
(Eccles & Midgley, 1989). Therefore, perceived social support at school
might be an especially critical factor that motivates students to pursue so-
cially adaptive goals at this time.
In addition, we know very little about how goals develop over the
course of the school years or about the relative contribution of schooling
and family socialization processes to individual differences in children's
adoption and pursuit of these goals. Whereas parents and teachers might
influence the internalization of goals and values in the early years, peers
might play an increasingly important role as children reach adolescence.
This might be especially true if peers become the primary source of emo-
tional support and nurturance as children make their way through middle
school and high school.
Continuity across contexts

An assumption drawn from research on family socialization processes is
that relationships characterized by authoritative parenting styles are en-
joyed by most children and that this style leads to the most adaptive out-
comes for all children. However, it is clear that many children do not have
relationships with parents that are authoritative in nature. Moreover, au-
thoritative parenting styles are not always related significantly to aspects
of children's school adjustment. This is especially true for children from
nonwhite, nonmiddle class families (Steinberg, Dornbusch, & Brown,
1992). Therefore, we must begin to explore the nature of similarities and
differences between home and school settings and carefully examine what
happens when children experience discontinuities between home and
school in the socialization processes and types of social supports they ex-
perience.
Social skills and self-regulatory processes

The adoption of social goals is dependent on several important social cog-
nitive factors. For instance, the pursuit of these goals reflects an ability to
perceive and understand social rules and conventions and to judge when
and where these rules are appropriate to follow. In addition, active pursuit
of socially integrative goals often requires an ability to recognize the exis-

tence of conflicting goals (those that a child wants to achieve and those
that a child is supposed to achieve) and then, to coordinate them in ways
that satisfy both personal and social objectives (Ford, Wentzel, Wood,
Stevens, & Seisfeld, 1989).
Reasons why conflicts between parents and children arise (Smetana,
1988), and ways in which these problems are resolved might also shed ad-
ditional light on young adolescents' classroom goal pursuit. Research on
how students perceive and consequently resolve conflicts between their
own personal goals and those that contribute to a positive instructional
classroom climate is especially needed in this regard.
Finally, although the literature reviewed in this chapter suggests that
the pursuit of social goals is related significantly and positively to social
as well as academic outcomes, it tells us little about the processes that link
these efforts to tangible results. Clearly, having good intentions is impor-
tant, but students must also be able to accomplish what they set out to
achieve. The processes by which goals are linked to social acceptance and
academic accomplishments are not well understood. Theoretical models
have suggested various psychological processes that regulate goal pursuit
and that mediate the initial adoption of goals and subsequent displays of
goal-directed behavior (e.g., Dodge, 1986; Ford, 1992). The next step is to
examine empirically the role of these self-regulatory and self-control
processes in linking social goal-setting to socially competent behavior in
the classroom.
Acknowledgment
I would like to thank Martin Ford for his constructive and insightful com-
ments on an earlier version of this chapter.
References
Allen, J. D. (1986). Classroom management: Students' perspectives, goals, and
strategies. American Educational Research Journal, 23, 437-459.
Baumrind, D. (1971). Current patterns of parental authority. Developmental Psy-
chology Monograph, 4, (I, Pt.2).
Baumrind, D. (1991). Effective parenting during the early adolescent transition. In
P. A. Cowan & M. Hetherington (Eds.), Family transitions (pp. 111-164). Hills-
dale, NJ: Erlbaum.
Blumenfeld, P. C, Hamilton, V L., Bossert, S. T., Wessels, K., & Meece, J. (1983).
Teacher talk and student thought: Socialization into the student role. In J. M.
Levine & M. C. Wang (Eds.), Teacher and student perceptions: Implications
for learning (pp. 143-192). Hillsdale, NJ: Erlbaum.
Burmester, D., & Furman, W. (1987). The development of companionship and in-
timacy. Child Development, 58, 1101-1113.
Cobb, J. A., & Hopps, H. (1973). Effects of academic survival skills training on
low achieving first graders. The Journal of Educational Research, 67, 108—
113.
Coie, J. D., & Krehbiel, G. (1984). Effects of academic tutoring on the social sta-
tus of low-achieving, socially rejected children. Child Development, 55,
1465-1478.
Sroufe (Eds.), Self processes and development: The Minnesota symposia on
child development (vol. 23, pp. 43-78). Hillsdale, NJ: Erlbaum.
Crandall, V J. (1966). Personality characteristics and social and achievement be-
haviors associated with children's social desirability response tendencies. Jour-
nal of Personality and Social Psychology, 4, 477-486.
Deci, E. L., & Ryan, R. M. (1991). A motivational approach to self: Integration in
personality. Nebraska symposium on motivation 1990 (pp. 237-288). Lincoln,
NE: University of Nebraska Press.
Dodge, K. A. (1986). A social information processing model of social compe-
tence in children. In M. Perlmutter (Ed.), Minnesota symposium on child psy-
chology (vol. 18, pp. 77-126). Hillsdale, NJ: Erlbaum.
Dreeben, R. (1968). On what is learned in school. Menlo Park, CA: Addison-
Wesley.
Dweck, C. S., & Leggett, E. L. (1988). A social-cognitive approach to motivation
and personality. Psychological Review, 95, 256-272.
Eccles (Parsons), J., Adler, T. E, Futterman, R., Goff, S. B., Kaczala, C. M,
Meece, J. L., & Midgley, C. (1983). Expectancies, values, and academic behav-
iors. In J. T. Spence (Ed.), Achievement and achievement motivation (pp.
75-146). San Francisco: W. H. Freeman.
Eccles, J., & Midgley, C. (1989). Stage-environment fit: Developmentally appro-
priate classrooms for young adolescents. In C. Ames & R. Ames (Eds.), Re-
search on motivation in education, (vol. 3, pp. 139-186). New York: Academic
Press.
Epperson, D. C. (1963). Some interpersonal and performance correlates of class-
room alienation. School Review, 71, 360-376.
Feldhusen, J. F, Thurston, J. R., & Benning, J. J. (1970). Longitudinal analyses of
classroom behavior and school achievement. Journal of Experimental Educa-
tion, 38, 4-10.
Feldlaufer, H., Midgley, C, & Eccles, J. S. (1988). Student, teacher, and observer
perceptions of the classroom before and after the transition to junior high
school. Journal ofEarly Adolescence, 8, 133—156.
Feldman, S. S., & Wentzel, K. R. (1990). The relationship between family interac-
tion patterns, classroom self-restraint, and academic achievement. Journal of
Ford, M. E. (1982). Social cognition and social competence in adolescence. De-
velopmental Psychology, 18, 323-340.
Ford, M. E. (1992). Motivating humans: Goals, emotions, and personal agency
beliefs. Newbury Park, CA: Sage.
Ford, M. E., & Nichols, C. W. (1991). Using goal assessments to identify motiva-
tional patterns and facilitate behavioral regulation and achievement. In M. L.
Maehr & P. Pintrich (Eds.), Advances in motivation and achievement (vol. 7,
pp. 51-84). Greenwich, CT: JAI.
Ford, M. E., Wentzel, K. R., Wood, D. N., Stevens, E., & Siesfeld, G. A. (1989).
search, 4, 405-425.
Goodenow, C. (1993). Classroom belonging among early adolescent students: Re-
lationships to motivation and achievement. Journal of Early Adolescence, 13,
21-43.
Grusec, J. E., & Goodnow, J. J. (1994). Impact of parental discipline methods on
the child's internalization of values: A reconceptualization of current points of
view. Developmental Psychology, 30, 4-19.
Hargreaves, D. H., Hester, S. K., & Mellor, F. J. (1975). Deviance in classrooms.
London: Routledge & Kegan Paul.
Hartup, W. W (1983). Peer relations. In P. H. Mussen (Ed.), Handbook of child
psychology, (vol. 4, pp. 104-196). New York: Wiley.
Hopps, H., & Cobb, J. A. (1974). Initial investigations into academic survival-skill
training, direct instruction, and first-grade achievement. Journal of Education-
al Psychology, 66, 548-553.
Horn, W. E, & Packard, T. (1985). Early identification of learning problems: A
meta-analy sis. Journal ofEducational Psychology, 77, 597-607.
Jackson, P. W. (1968). Life in classrooms. New York: Holt, Rinehart & Winston.
Kozeki, B. (1985). Motives and motivational style in education. In N. Entwistle
(Ed.), New directions in educational psychology: 1. Learning and teaching (pp.
189-199). Philadelphia: Falmer Press.
Kozeki, B., & Entwistle, N. J. (1984). Identifying dimensions of school motiva-
tion in Britain and Hungary. British Journal of Educational Psychology, 54,
306-319.
Ladd, G. W, & Price, J. M. (1987). Predicting children's social and school adjust-
ment's, 1168-1189.
Lambert, N. M. (1972). Intellectual and non-intellectual predictors of high school

status. Journal of Special Education, 6, 247—259.
Maccoby, E. E. (1992). Trends in the study of socialization: Is there a Lewinian
heritage? Journal of Social Issues, 48, 171-185.
Maccoby, E. E., & Martin, J. A. (1983). Socialization in the context of the family:
Parent-child interaction. In P. H. Mussen (Ed.), Handbook of child psychology,
(vol. 4, pp. 1-101). New York: Wiley.
Maehr, M. L. (1983). On doing well in science: Why Johnny no longer excels; why
Sarah never did. In S. G. Paris, G. M. Olson, & H. W. Stevenson (Eds.), Learn-
ing and motivation in the classroom (pp. 179-210). Hillsdale, NJ: Erlbaum.
Midgley, C, Feldlaufer, H., & Eccles, J. (1989). Student/teacher relations and atti-
tudes toward mathematics before and after the transition to junior high school.
Miller, N. (1983). Peer relations in desegrated schools. In J. L. Epstein & N. Kar-
weit (Eds.), Friends in school (pp. 201-217). New York: Academic Press.
Nakamura, C. Y, & Finck, D. N. (1980). Relative effectiveness of socially orient-
ed and task-oriented children and predictability of their behaviors. Monographs
of the Society for Research in Child Development, 45, (Nos. 3~4).
Nicholls, J. G. (1984). Achievement motivation: Conceptions of ability, subjective
experience, task choice, and performance. Psychological Review, 91, 328-346.
Ogbu, J. U. (1985). Origins of human competence: A cultural-ecological perspec-
tive. Child Development, 52, 413^29.
Pervin, L. A. (1982). The stasis and flow of behavior: Toward a theory of goals. In
M. M. Page (Ed.), Personality-Current theory and research (pp. 1-53). Lin-
coln, NE: University of Nebraska Press.
Reid, M., Landesman, S., Treder, R., & Jaccard, J. (1989). "My family and
friends": Six- to twelve-year-old children's perceptions of social support. Child
Safer, D. J. (1986). Nonpromotion correlates and outcomes at different grade lev-
els. Journal of Learning Disabilities, 19, 500-503.
Schunk, D. H., & Zimmerman, B. J. (1996). Modeling and self-efficacy influ-
ences on children's development of self-regulation. In J. Juvonen & K. R.
Wentzel (Es.), Social motivation: Understanding children s school adjustment.
pp. 154-180. New York: Cambridge.
Shultz, J., & Florio, S. (1979). Stop and freeze: The negotiation of social and
physical space in a kindergarten/first grade classroom. Anthropology and Edu-
cation Quarterly, 10, 166-181.
Sieber, R. T. (1979). Classmates as workmates: Informal peer activity in the ele-
mentary school. Anthropology and Education Quarterly, 10, 207-235.
Smetana, J. G. (1988). Concepts of self and social convention: Adolescents' and
parents' reasoning about hypothetical and actual family conflicts. In M. R.
Gunnar (Ed.), Minnesota symposium on child psychology (pp. 79-122). Hills-
dale, NJ: Erlbaum.
Steinberg, L., Dornbusch, S. M., & Brown, B. B. (1992). Ethnic differences in
adolescent achievement: An ecological perspective. American Psychologist, 47,
723-729.
tion. In C. P. Smith (Ed.), Achievement-related motives in children (pp.
46-101). New York: Russell Sage.
Weiner, B. (1985). An attribution theory of achievement motivation and emotion.
and academic achievement: An interactionist perspective. Journal of Educa-
tional Psychology 81, 131-142.
Wentzel, K. R. (1991a). Relations between social competence and academic
Wentzel, K. R. (1991b). Social competence at school: The relation between social
responsibility and academic achievement. Review of Educational Research, 61,
1-24.
Wentzel, K. R. (1991c). Social and academic goals at school: Achievement moti-
vation in context. In M. Maehr & R Pintrich (Eds.), Advances in motivation and
achievement (vol. 7, pp. 185-212). Greenwich, CT: JAI.
Wentzel, K. R. (1992). Motivation and achievement in adolescence: A multiple
goals perspective. In J. Meece & D. Schunk (Eds.), Student perceptions in the
classroom: Causes and consequences (pp. 287-306). Hillsdale, N. J.: Erlbaum.
Wentzel, K. R. (1993a). Social and academic goals at school: Motivation and
achievement in early adolescence. Journal of Early Adolescence, 13, 4-20.
Wentzel, K. R. (1993b). Does being good make the grade? Relations between aca-
demic and social competence in early adolescence. Journal ofEducational Psy-
chology, 85, 357-364.
Wentzel, K. R. (1994a). Family functioning and academic achievement in middle
school: A social-emotional perspective. Journal of Early Adolescence, 14,
268-291.
Wentzel, K. R. (1994b). Relations of social goal pursuit to social acceptance,
classroom behavior, and perceived social support. Journal of Educational Psy-
chology, 86, 173-182.
Wentzel, K. R. (1996). Social Support and Achievement: The Role of Parents,
Teachers, and Peers. Unpublished manuscript, University of Maryland.
Wentzel, K. R., & Asher, S. R. (1995). Academic lives of neglected, rejected, pop-
ular, and controversial children. Child Development, 66, 754-763.
Wentzel, K. R., & Feldman, S. S. (1993). Parental predictors of boys' self-restraint

and motivation to achieve at school: A longitudinal study. Journal ofEarly Ado-
lescence, 13, 183-203.
Wentzel, K. R., Weinberger, D. A., Ford, M. E., & Feldman, S. S. (1990). Academ-
ic achievement in preadolescence: The role of motivational, affective, and self-
regulatory processes. Journal of Applied Developmental Psychology, 11,
179-193.
Youniss, I, & Smollar, J. (1989). Adolescents' interpersonal relationships in so-
cial context. In T. Berndt & G. Ladd (Eds.), Peer relationships in child develop-
ment (pp. 300-316). New York: Wiley.
11 Friends9 influence on school adjustment:
A motivational analysis
Thomas J. Berndt and Keunho Keefe
Best friends can have a powerful influence on children's attitudes toward

school, behavior in class, and academic achievement. The influence of
friends has long been a concern of educators and educational researchers.
Several decades ago, James Coleman (1961) argued that most high-school
students care more about being popular with peers than about doing well
in school, in part because their peers emphasize academic success less
than social success. Many recent writers have echoed this theme (e.g.,
Bishop, 1989).
By contrast, other writers have emphasized the positive effects of
friendships on children's adjustment and development. Piaget (1932/
1965) proposed that interactions with friends or other peers are crucial for
the development of a mature morality. Sullivan (1953) suggested that inti-
mate friendships among preadolescents contribute to high self-esteem and
to social understanding (Buhrmester & Furman, 1986). More recently,
many researchers have tested the hypothesis that support from friends en-
hances the social and academic adjustment of children and adolescents
(e.g., Berndt & Keefe, 1995).
Both perspectives on friends' influence capture part of the truth, but
both are incomplete and therefore misleading. Friends influence children
and adolescents through two distinct pathways (Berndt, 1992). First, stu-
dents at all grade levels are influenced by the attitudes, behavior, and oth-
er characteristics of their friends. This influence is not always negative.
Students whose friends have positive characteristics, such as high grades,
are likely to improve their own grades over time (Epstein, 1983). One goal
of our chapter is to outline various motives that account for friends' influ-
ence by this pathway.
Second, students are influenced by the quality of their friendships. For
example, friendships are higher in quality when they are more intimate
(Sullivan, 1953). Friendships are lower in quality when the friends often
248
Friends' influence on school adjustment 249
engage in conflicts or hostile rivalry with one another (Laursen, 1993).

The quality of students' friendships affects their social and emotional ad-
justment. It can, in turn, affect their adjustment to school. The second goal
of our chapter is to outline motives related to friendship quality. Identify-
ing these motives should increase understanding of this second pathway
of influence.
Before proceeding, we should say more precisely what we mean by mo-
tives and by school adjustment. Weiner (1992) proposed that the domain of
motivation includes all questions about why organisms think and behave as
they do. This broad definition includes motives that actors themselves re-
port. For example, students sometimes say that their goal at school is trying
to do what a teacher asks (Wentzel, 1991a). The definition also includes
motives of which a person may not be aware. Unconscious motives are em-
phasized not only in psychoanalytic theories, but also in some learning the-
ories and social-psychological theories (see Nisbett & Wilson, 1977; Wein-
er, 1992). Both types of motives are considered in the chapter.
Like motivation, school adjustment is a broad construct. Our focus is
on three facets of adjustment. First, students' adjustment to school is re-
flected by their attitudes toward their classes, their teachers, and other ex-
periences at school. Well-adjusted students value what they are learning
and are positively involved in classroom activities (Berndt & Miller, 1990;
Wentzel, 1993). Second, students' adjustment is reflected by their class-
room behavior. Well-adjusted students behave appropriately and are rarely
disruptive (Berndt & Keefe, in press; Dubow, Tisak, Causey, Hryshko, &
Reid, 1991). Third, students' adjustment is reflected by their academic
achievement. Well-adjusted students learn what is taught in school, and so
receive high grades and test scores.
Friends may affect all three facets of school adjustment through both
influence pathways outlined earlier. The first pathway is the focus of the
next section of the chapter. We emphasize that many motives contribute to
the influence of friends' characteristics. The second pathway is the focus
of the following section. We emphasize that friends have goals not only
for themselves as individuals, but for themselves and their friends viewed
as a unit. An outline of the two pathways of influence and their associated
motives is given in Table 11.1.
Also discussed in the initial sections is how friends' influence through
the two pathways might change with age. Unfortunately, little evidence on
such changes is available, and the evidence is not always consistent. We
summarize the conclusions that can be drawn, and then use our motiva-
tional analysis to suggest new perspectives on the issue.
Table 11.1. Motives contributing to friends' influence through two pathways

Pathway I: Influences of
friends' characteristics Pathway II: Effects of friendship quality
1. The need for approval 1. Relationship ("we") motives (related to positive features)
2. Identification with friends 2. Individualistic/competitive motives (related to negative
3. The self-enhancement motive features)
4. The need to be correct
In a third section, we consider implications of the motivational analysis

for educational practice. We focus specifically on strategies for dealing
with friends who negatively influence one another. Our recommended
strategies must be interpreted cautiously, because many questions about
friends' influence remain open. In the final section, we suggest directions
for future research on these questions.
Motives related to the influence of friends' characteristics

Both popular and scholarly writers have suggested that pressure from
peers contributes to drug use, delinquent behavior, and apathy toward
school (Ansley & McCleary, 1992; Bishop, 1989; Steinberg & Silverberg,
1986). These writers rarely say which peers are the presumed source of
this negative pressure, but other data suggest that close friends have more
influence than do other peers (Downs, 1987). Nevertheless, many kinds of
evidence show that this view of friends' influence is seriously distorted.
We begin this section by examining evidence related to the notion of
friends' pressure. Then we consider four motives that better explain why
students' friends affect their adjustment to school. Finally, we review the
few existing studies of developmental changes in this pathway of influ-
ence.
Inaccurate ideas about friends' pressure

The notion that pressure from friends has strong and negative effects on
children and adolescents is inaccurate in three major ways. First, this no-
tion exaggerates the magnitude of friends' influence. A popular maga-
zine recently published a survey that was completed by more than
100,000 adolescents (Ansley & McCleary, 1992). One question asked,
"How much peer pressure are you under to do things that probably are
wrong?" Although the reporters who summarized the results concluded
that "peer pressure is a big problem," the adolescents' responses suggest
otherwise. Most adolescents (66%) said the amount of pressure they

were under was "not much." Another 14% said they were under no pres-
sure to do things that are wrong. Only 20% of the adolescents said they
were under "a lot" of pressure.
More systematic research confirms that strong pressure from friends to
engage in specific behaviors is rare. Naturalistic observations of adoles-
cents' interactions with friends have shown that most decisions are
reached by consensus rather than coercive pressure (Suttles, 1968). Usual-
ly, the friends discuss various options until they agree on what to do.
Rarely is one adolescent pressured to conform to the rest of the group.
Research on friends' pressure and school adjustment is limited. Cole-
man's (1961) evidence for negative effects of peers on academic achieve-
ment was indirect, and writers such as Bishop (1989) mentioned no em-
pirical data at all. We obtained relevant data with two samples of
junior-high-school students (Berndt, Miller, & Park, 1989). We asked stu-
dents how much influence their friends had on the three aspects of school
adjustment defined earlier: attitudes, behavior in class, and academic
achievement. Most students said their friends did not affect these aspects
of their school adjustment. Evidence presented later implies that the stu-
dents underestimated their friends' influence on them, but the students'
comments suggest that any influence of friends did not result from overt
pressure.
A second inaccuracy in popular ideas about friends' pressure concerns
the direction of its effects. For most students, these effects are more likely
to be positive than negative. High-school students in several large samples
reported that their friends discouraged drug and alcohol use, delinquent
activities, and other types of antisocial behavior more than they encour-
aged them (Brown, Clasen, & Eicher, 1986; Keefe, 1994). The students
also said their friends encouraged studying for school subjects more than
they discouraged studying.
The third inaccuracy in popular ideas about friends' pressure concerns
the motives that account for friends' influence. These ideas imply that stu-
dents conform to their friends because they fear they will be punished if
they don't. However, students can escape any punishment by ending those
friendships and making new friends. Furthermore, neither children nor
adolescents are as lacking in independence as is often suggested. When
completing surveys, many adolescents report that they make their own de-
cisions rather than relying on the advice of parents or friends (Sebald,
1986). In experimental studies, some children and adolescents show anti-
conformity, rejecting their peers' judgments and making different ones
(Hartup, 1983).
According to Piaget (1932/1965), friendships have features that limit

both the use and the effectiveness of coercive pressure. Friends assume
that their relationship is based on equality and mutual respect, so deci-
sions must be made through negotiation rather than domination by one
person. Adolescents in our previous study suggested this kind of mutual
respect when explaining why their friends had no influence on their
school adjustment (Berndt et al., 1989). For example, one boy said that his
friend "doesn't try to change my opinions, and I don't try to change his."
This argument should not be carried too far. Although students' adjust-
ment to school does not depend greatly on their response to friends' pres-
sure, friends do influence attitudes, behavior, and achievement in school
(Berndt & Keefe, in press; Epstein, 1983; Kandel, 1978). By drawing
upon a long tradition of developmental and social-psychological research,
we can identify the motivation that accounts for these effects of friends'
characteristics.
Four motives that underlie friends' influence. For decades, questions

about interpersonal influence have been explored by social psychologists
interested in persuasion, attitude change, decision making in groups, and
other phenomena (Chaiken, Liberman, & Eagly, 1989; Zanna, Olson, &
Herman, 1987; Zimbardo & Leippe, 1991). These questions have also
been explored by developmental psychologists interested in social learn-
ing, conformity to peers and parents, basic socialization processes, and
other topics in social development (Bandura & Walters, 1963; Hartup,
1983; Maccoby & Martin, 1983).
Not surprisingly, a strong consensus about the motives responsible for
interpersonal influence does not exist. The lack of consensus is due partly
to researchers' use of different terms to refer to the same or overlapping
constructs. In addition, multiple motives are linked to social influence,
and many researchers have focused on only one. We argue that four mo-
tives should be considered when trying to understand how students are af-
fected by their friends' adjustment to school (see Table 1). Some writers
might propose motives other than the four that we present, but we assume
other motives are less important than the following four or are largely syn-
onymous with them (see, however, Erdley; Ford; Kupersmidt et al.,
Wentzel, this volume, for alternative perspectives on social goals).
The first motive that underlies the influence of friends' characteristics
is students' need for social approval (Hoving, Hamm, & Galvin, 1969; Ju-
vonen & Weiner, 1993) or their impression motivation (Chaiken et al.,
1989). Students want to be liked by their friends, so they try to do things
that will meet the friends' expectations or make a positive impression on

them. This motive is associated theoretically with the idea of social rein-
forcement. Praise and other kinds of positive comments from friends can
function as rewards for specific behaviors by students and so increase the
likelihood of those behaviors (Hartup, 1983). In the broadest sense, this
motive relates to Skinner's principle of positive reinforcement, a principle
common to all learning theories.
The category of positive reinforcement includes both material rewards
and social reinforcers like praise. However, for understanding the influ-
ence of friends, praise and other social reinforcers are more important
than are material rewards. Friends rarely give things to one another as re-
wards for specific behaviors. Therefore, the construct of social approval
corresponds to the attributes of friends' interactions more precisely than
does the broader construct of positive reinforcement.
In the domain of school adjustment, the importance of the needs for
social approval was first emphasized by Crandall, Katkovsky, and Preston
(1960). They argued that "the basic goal of achievement behavior is the at-
tainment of approval and the avoidance of disapproval" (p. 791). They as-
sumed that approval comes from both teachers and peers, but little of their
research focused directly on this motive. In particular, they did not explore
whether the need for approval is related to friends' influence on school ad-
justment.
Many studies in laboratory settings have shown the effects of peer rein-
forcement on children's behavior (Hartup, 1983). Therefore, it is reason-
able to assume that reinforcement from friends and other classmates also
affects students' attitudes, behavior, and achievement in school. Whether
friends typically encourage a positive or a negative adjustment to school is
a controversial question. Consistent with the positive view is the evidence
mentioned earlier that students say friends encourage them to study hard
at school (Brown et al., 1986). Also consistent with this view is evidence
that students higher in academic achievement are usually more popular
with peers (Coie, 1990).
There is conflicting evidence, however. The eighth graders in one re-
cent study (Juvonen & Murdock, 1993; see also Juvonen, this volume)
said they thought a peer who got good grades would generally be more
popular than a peer who got bad grades. However, the eighth graders
thought a peer who got bad grades but who was very smart and tried very
hard would be more popular than a peer with the same attributes who got
good grades. To explain this finding, the researchers speculated that the
eighth graders might have assumed the high-achieving student would "set
the curve" for the class and make them look worse. Alternatively, the
eighth graders might have viewed the high achiever as accepting tradition-
al expectations about the value of school achievement that they them-
selves rejected. Notice that these two explanations differ in that they sug-
gest that students generally value, or are disdainful of, high academic
achievement.
Other findings indicate that students' level of academic achievement
affects their selection of friends. Friendship selection, in turn, affects the
attitudes and behavior for which students receive friends' praise (Ball,
1981; Schwartz, 1981). High-achieving students are usually friends with
classmates who encourage them to get good grades and not to misbehave
in class. Low-achieving students are usually friends with classmates who
express no interest in getting good grades and who often misbehave in
class. Moreover, students shift over time toward their friends' attitudes and
behavior. The shift can be attributed partly to the students' need for their
friends' approval.
A second motive that partly accounts for the influence of friends is the
desire of students to think and behave like their friends. Students normally
choose best friends who have characteristics or talents that the students
admire (Hallinan, 1983). This admiration motivates students to act as their
friends do.
Kelman (1961) described the process leading to emulation of an ad-
mired individual as identification. The concept of identification originat-
ed in Freudian theory, where it had multiple meanings (Mischel, 1970).
Identification was used to refer to actual imitation of another's behavior,
to the mechanisms leading to imitation, to a motive to be like another, and
to a belief that one has the same attributes as the other. Later, social learn-
ing theorists argued that the core meanings of identification were the
same as their principles of observational learning and imitation (Bandura,
1969). That is, people learn how and when to perform certain behaviors by
observing other people.
Observational learning is different from Kelman's (1961) concept of
identification, however. Social learning theorists assume that people imi-
tate others' behavior mainly because they believe they will be rewarded
for doing so (Bandura, 1977). By contrast, Kelman retained Freud's as-
sumption that identification depends on a specific motive to be like the
other person. It depends, therefore, on having a special relationship with
the other person.
Kelman also argued that identification involves a continuing relation-
ship between the person being influenced and the person who is the source
of influence. One byproduct of this continuing relationship is that the be-

havior of the person being influenced changes whenever the other person's
behavior changes. For example, when students are strongly identified with
particular friends, they will change their own interests and activities
whenever those of their friends change. When the friends express dislike
for a particular teacher, the students will also form negative attitudes to-
ward that teacher. If the friends later begin to like that teacher for some
reason, the students will also develop greater liking for the teacher. By
contrast, the principle of observational learning does not imply that stu-
dents will track the variations in their friends' behavior so closely, because
it does not assume such enduring links between models and observers.
Several experimental studies of observational learning from peers have
been done (Hartup, 1983). Few studies, however, have focused on aspects
of school adjustment or examined the influence of friends in particular.
The effects of peer models on students' achievement and self-efficacy
have been demonstrated (e.g., Schunk, Hanson, & Cox, 1987), but these
studies did not use friends as models. In addition, friends' similarity in
academic achievement and educational aspirations has often been attrib-
uted to modeling (e.g., Kandel & Andrews, 1987). However, the processes
leading to friends' similarity have not been assessed directly.
Despite the absence of direct evidence, few researchers or educators
would deny that students' identification with friends has some effect on
their school adjustment. Students' desires to behave exactly like their
friends may often be exaggerated, but students certainly pay attention to
how their friends talk about school, behave in school, and achieve acade-
mically.
Moreover, the motive to identify with friends is theoretically signifi-
cant because it falls within the category of intrinsic motivation. Students
do as their friends do, not because their friends provide rewards for imita-
tion, but because the friends have positive characteristics that the students
want to have. Thus, this motive contrasts sharply with the notion that
friends have influence because they punish students who try to resist their
coercive pressure.
A third motive related to friends' influence is self-enhancement. Stu-
dents partly judge their own competence by comparing their performance
with that of their classmates. According to Veroff (1969), social compari-
son increases students' motivation to achieve and contributes to aggressive
competition with classmates. VerofF proposed that students who are the
winners in academic competition often receive approval from peers as well.
This social approval further enhances their social comparison motivation.
Veroff's ideas are intriguing because they contrast sharply with the
more recent speculations of Juvonen and Murdock (1993; see also Juvo-
nen, this volume). Recall that these researchers assumed competition in
academics is resented by classmates and reduces a student's popularity.
These opposing viewpoints also exist when the question is focused on
friends rather than all peers. As noted earlier, friendships are based on
equality, which makes friends likely targets for social comparison
(Berndt, 1986). Veroff's (1969) theory implies that these comparisons en-
hance students' efforts to achieve academically, to prove they are as good
as, or better than, their friends. Juvonen and Weiner's (1993) ideas imply
that these comparisons make students less eager to achieve academically,
because their success might make their friends resentful.
Tesser's (1984) self-esteem maintenance model suggests a resolution to
this controversy. He argues that students try to show their superiority to
friends in areas most relevant to their self-esteem, while admitting their
friends' superiority to them in less relevant areas. Thus, fifth and sixth
graders who consider mathematics as very important also rate their per-
formance in math more highly than that of their friends. Fifth and sixth
graders who consider mathematics as relatively unimportant rate their
friends' performance in math more highly than their own (Tesser, Camp-
bell, and Smith, 1984).
However, students are not completely free to define domains of
achievements as unimportant. For example, not all parents would accept
their child's assertion that getting a "C" in math is unimportant. For this
reason, students cannot always avoid direct comparisons with friends. In
addition, they cannot always escape the academic competition that such
comparisons provoke.
Ethnographic research suggests that friends' competition takes differ-
ent forms, depending on a student's level of achievement. High-achieving
students often compete with friends to get the best grades on tests (Ball,
1981; Schwartz, 1981). These students express the self-enhancement mo-
tive by trying to show their academic superiority to friends. Low-achiev-
ing students often compete with friends in misbehavior, trying to enhance
their self-esteem by creating the greatest disruption in class or by chal-
lenging teachers' authority most directly (Ball, 1981; Schwartz, 1981). In
short, the motive to distinguish oneself- as a scholar or as a scoundrel -
can partly explain both students' own behavior and a route by which
friends influence them.
The fourth motive that partly accounts for friends' influence is the need
to be correct (Hoving et al., 1969), or validity-seeking (Chaiken et al.,
1989). This motive refers to a person's desire to hold correct beliefs and
make reasonable decisions. Deutsch and Gerard (1955) suggested the dis-
tinction between this motive and the three previous ones when they con-
trasted processes of normative and informational influence.
Normative influence depends on a person's desire to conform to the
positive expectations of others. Viewed narrowly, it is the same as the need
for social approval. This category has been expanded to include all influ-
ence processes that involve a person's reactions to other people's opinions
and behavior (Isenberg, 1986). This broad definition encompasses all
three motives discussed earlier.
By contrast, informational influence depends on a person's acceptance
of another person's arguments as evidence about reality (Deutsch & Ger-
ard, 1955). In other words, the person focuses on the accuracy of those ar-
guments rather than their source. This category has been expanded in
more recent research to include all influence processes that involve a per-
son's comprehension and evaluation of relevant arguments (Isenberg,
1986).
Many types of studies show that the need to be correct provides a par-
tial explanation of friends' influence on children and adolescents. Its im-
portance is perhaps most obvious in research on peer collaboration during
problem solving (Damon & Phelps, 1989). Students often show improved
performance on cognitive problems after working on them with a peer.
This improvement is not due entirely to observational learning, that is,
poorer students learning from better students. Often, two students who are
working together both gain a better understanding of cognitive problems
after listening to one another's arguments (Tudge, 1992).
Studies of peer collaboration can partly explain the influence of friends
on one another's academic achievement (Epstein, 1983). This influence
must depend partly on the friends' collaboration on class work or home-
work assignments. But informational influence goes beyond academic
work itself. A need to be correct, or to hold valid opinions, also provides a
partial explanation of friends' influence on attitudes about school.
In a recent study (Berndt, Laychak, & Park, 1990), junior high school
students made decisions on hypothetical dilemmas that pitted doing school
work against social activities or more free time. For example, on one dilem-
ma students had to decide whether to go to a rock concert one evening or to
stay home and study for an exam. After making decisions independently,
students in one condition discussed the decisions with a close friend and
tried to agree on them. In another condition, students discussed topics un-
related to school, such as where to go on a summer vacation.
After the discussions, the students again made decisions on the dilem-
mas independently. The independent decisions of friends who had dis-
cussed the dilemmas were more similar than those of friends who had not,
showing that friends influenced one another's decisions. Students also
shifted after the discussions toward the decisions that were accompanied
by the most reasons during the discussions. These findings imply that dur-
ing adolescence, friends' influence depends on information exchange and
the motive to be correct, just as is true in adulthood (Chaiken et al., 1989;
Isenberg, 1986).
Age changes in the influence of friends

Does the influence of friends' characteristics change with age? To answer
this question, a few researchers asked students to read about hypothetical
situations in which friends supposedly encouraged them to engage in spe-
cific behaviors (Berndt, 1979; Brown et al., 1986; Steinberg & Silverberg,
1986). Then students reported whether they would conform to the friends'
suggestions. Most often, apparent conformity to friends increased be-
tween middle childhood and mid-adolescence, or around ninth grade.
Conformity then decreased between mid-adolescence and late adoles-
cence, or the end of high school. However, this developmental trend was
not significant in all samples (Brown et al., 1986).
Other approaches to the estimation of friends' influence have also
yielded mixed results. A few researchers have tried to assess the increase
in friends' similarity that results from their influence on each other. In one
study (Urberg, Cheng, & Shyu, 1991), friends seemed to influence eighth-
graders' cigarette smoking more than that of eleventh graders. In another
study (Epstein, 1983), friends' influence on school-related attitudes and
on academic achievement seems to change little between fourth and
twelfth grade.
Another way to explore the question of developmental changes would
be to examine the strength of the various motives that underlie friends' in-
fluence. Unfortunately, this approach has not often been adopted. Re-
searchers once assumed that students' need for social approval could be
estimated from their conformity to peers on certain types of judgments.
However, the available data on peer conformity are difficult to interpret
(Hartup, 1983). Our second motive, to identify with friends or want to be
like them, has rarely been examined directly. Furthermore, few re-
searchers have examined age changes in the related process of observa-
tional learning from peers. Therefore, it is impossible to say whether this
motive changes in strength as children grow older.
The self-enhancement motive may become more important during

middle childhood. A few studies suggest that children's self-evaluations
are more strongly affected by social comparisons with peers with increas-
ing age (e.g., Ruble, Boggiano, Feldman, & Loebl, 1980). This age trend
is consistent with evidence that self-esteem decreases during middle
childhood, as children better appreciate how their performance in various
domains compares with that of their peers (Marsh, 1989). Veroff (1969)
suggested that this trend is partly reversed in early adolescence, as social
comparison is integrated with autonomous motives for achievement. Un-
fortunately, little research has been done on this hypothesis.
Finally, some data suggest that group discussions are more rational, or
based more on informational influence, as children move into adolescence
(Berndt, McCartney, Caparulo, & Moore, 1983-1984; Smith, 1973). Ado-
lescents give more reasons for their opinions than children; they also re-
solve conflicts during discussions more effectively.
More research on the age changes in motives linked to friends' influ-
ence would be valuable. Nevertheless, it is important to recognize that
friends can have a strong influence on certain aspects of school adjust-
ment throughout the school years. Even in the elementary grades, stu-
dents' behavior is affected by the disruptive behavior of their friends
(Schwartz, 1981). Near the end of high school, students' educational aspi-
rations are affected by their friends' aspirations (Davies & Kandel, 1981).
More research is needed on how much friends affect each aspect of school
adjustment during each phase of schooling.
Motives related to the effects of friendship quality

To understand the effects of friendship quality on students' adjustment,
the construct of friendship quality must first be defined. In this section,
we provide a definition that links features of friendship to their associated
motives. Then we review research that indicates the effects of friendships
on school adjustment. Next, we consider the limited research on develop-
mental changes in the effects of friendships.
High-quality friendships: Features and motives

Piaget (1932/1965) and Sullivan (1953) assumed that friendships are high
in quality when they are intimate, egalitarian, and based on mutual re-
spect. To supplement these definitions, other writers have suggested that
high-quality friendships are high in prosocial behavior (sharing and help-
ing), trust, loyalty, affection, companionship, and caring (Berndt & Perry,
1986; Bukowski, Hoza, & Boivin, 1994; Furman & Buhrmester, 1992;
Parker & Asher, 1993).
Few writers have explicitly discussed the motives associated with high-
quality friendships. However, Sullivan's analysis of friendship included a
few suggestions about these motives. He referred to a close friend as a
chum, and suggested that children who have a chum say to themselves,
"What should I do to contribute to the happiness or to support the prestige
and feeling of worth-whileness of my chum?" (p. 245). That is, their mo-
tive is to make the friend happy and boost the friend's self-esteem.
As a complement to this apparently altruistic motive, Sullivan suggest-
ed that high-quality friendships fulfill a need for interpersonal intimacy.
During intimate conversations, friends share their concerns and are as-
sured that they are respected by peers whom they also respect. Sullivan
implied, however, that high-quality friendships do not involve a combina-
tion of altruistic and self-interested motives. The goal of friends' interac-
tion is collaboration, or "the pursuit of increasingly identical - that is,
more and more nearly mutual - satisfactions" (p. 246).
When friends aim for mutually satisfying interactions, they reject the
distinction between the goals of self and of friend. In this sense, the pri-
mary motive of persons involved in high-quality friendships is qualitative-
ly different from the individualistic motives (e.g., the need for approval)
considered earlier (see Table 11.1). In high-quality friendships, these indi-
vidualistic motives are replaced by motives for both partners in the rela-
tionship. They focus not on what "I" want, but what "we" want (Hartup,
1992).
Of course, actual friendships fall short of the complete mutuality de-
scribed by Sullivan. Aristotle (Ostwald, 1962) said that some people want
friends who increase their own pleasure. Other people want friends who
are useful to them, who will do favors for them and help them make valu-
able contacts with other people. Aristotle believed that perfect friendships,
which have the kind of intimacy and mutuality described by Sullivan, are
extremely rare.
Nevertheless, Aristotle assumed that most close friendships have some
features of a perfect friendship. Researchers who assess the positive fea-
tures of students' friendships are exploring how well these friendships
match the ideal in classical and modern writings. That is, they are examin-
ing the degree to which friends adopt relationship motives, thinking not of
"me" and "you" but of "us."
Friendships also have negative features. Earlier, we mentioned the
forces that lead to competition between friends (Tesser, 1984). Students
often compare their performance in academic and other activities to that

of their friends. For example, friends often compare their grades on tests
in school and take the results as an indicator of who is the smartest. Be-
cause their self-esteem is at stake, friends sometimes compete more in-
tensely with one another in such situations than nonfriends do (Berndt,
1986; Berndt, Hawkins, & Hoyle, 1986).
Competition between friends may not only affect each friend's behav-
ior, as discussed earlier. This competition may also affect the quality of
their friendship. Recall that Piaget (1932/1965) assumed peer relation-
ships are based on equality. Friends, especially, view themselves as equal
in all important respects (Berndt, 1986; Youniss & Smollar, 1985). But
when students are high in competitive motivation, they may reject this
view and try to prove their superiority to friends. Then the quality of their
friendships is likely to suffer.
Other types of conflicts between friends arise for various reasons
(Laursen, 1993). The root of many conflicts may be students' emphasis on
individualistic goals (Asher & Renshaw, 1981; Putallaz & Sheppard,
1990). Stated informally, some students would rather get what they want
and lose a friend than vice versa. Measures of negative friendship features
reflect the degree to which friends favor competitive and individualistic
goals over the goal of mutually satisfying outcomes.
Negative interactions between friends are less common in friendships
with many positive features, but the correlation is weak. Especially as
children move into adolescence, their reports about the positive and nega-
tive features of their friendships become more independent (Berndt &
Keefe, in press; Berndt & Perry, 1986). Therefore, the quality of a friend-
ship should be judged from separate assessments of its positive and nega-
tive features. The effects of the two aspects of friendship quality on school
adjustment should also be judged separately.
Effects offriendship quality on school adjustment

Children and adolescents whose friendships have more positive features
are higher in self-esteem and prosocial behavior, are more popular with
peers, and less often suffer from emotional problems (Berndt & Savin-
Williams, 1993; Hartup, 1992). They also have more positive attitudes to-
ward school, are better behaved, and are higher in academic achievement
than other students (Berndt & Keefe, in press; Dubow et al., 1991; Kurdek
& Sinclair, 1988). In addition, students whose friendships have more neg-
ative features report less classroom involvement and more disruptive be-
havior. By contrast, the number of best friends that students report usually
is only weakly correlated with their social and school adjustment (Savin-
Williams & Berndt, 1990). In research with adults, measures of number of
friendships are usually less strongly related to indicators of psychological
health than are measures of the support those relationships provide (Sara-
son, Sarason, & Pierce, 1990). In sum, relationship quality matters more
than quantity.
Correlations between friendship features and school adjustment do not
prove that friendship quality affects school adjustment. Students' adjust-
ment to school could instead affect the quality of the friendships they can
form. To distinguish between these alternatives, a few researchers have
examined the relations of friendship quality to the changes over time in
students' adjustment. In one recent study (DuBois, Felner, Brand, Adan, &
Evans, 1992), a measure of support from friends was not related to the
changes over two years in adolescents' grades. In a second study (Dubow
et al., 1991), friends' support was not related to the changes over two years
in younger students' behavioral and academic adjustment.
The two studies might have yielded null results because the interval be-
tween assessments was too long. Best friendships among children and
adolescents usually last for several months, but not for years (Hallinan,
1978/1979). With an interval of two years, researchers may have been try-
ing to assess effects of friendships that ended months before.
Another possible explanation for the null results is that the measures of
friends' support were too general. In one study (Dubow et al., 1991), the
items referred to support from classmates as well as friends. In the other
study (DuBois et al., 1992), the items referred to friends but not specifi-
cally to best friends. Both measures included items about positive features
but not negative ones.
Recently, we completed a longitudinal study that was less subject to
these problems (Berndt & Keefe, in press). In the fall of a school year, ju-
nior-high-school students described the positive and negative features of
their three best friendships. The students also reported their involvement
in classroom activities and their disruptive behavior at school. Teachers
rated the students on their involvement and disruptive behavior, and re-
ported their report-card grades. These assessments were repeated in the
following spring, about six months later. The data were analyzed in a hier-
archical regression analysis that took into account the continuity in school
adjustment. Therefore, the results can be interpreted as evidence regarding
the effects of friendship quality on school adjustment.
Students' reports on the positive features of their very best (or closest)
friendship were related to the changes during their year in their self-re-
Table 11.2. Predicting students' adjustmentfrom the features of their friend-

ships
Very best friend Multiple friends
2 2 2
R R changej Beta R R2 change Beta
Time 2 involvement (self-reported)
Step 1: Time 1 involvement .384 .384 .62*** .385 .385 .62***
Step 2: Time 1 Friendship Features
Positive features .393 .009 .09* .390 .005 .08
Negative features .388 .004 -.06 .389 .004 -.06
Time 2 disruption (self-reported)
Step 1: Time 1 disruption .389 .389 .62*** .389 .389 .62***
Step 2: Time 1 Friendship Features
Positive features .391 .002 -.04 .389 .000 -.01
Negative features .407 .018 .13** .410 .021 .15**
For each Time 2 measure of adjustment, the values listed in Step 2 are from two separate analy-
ses, one including only the variable for positive features and the other including only the variable
for negative features. For all analyses, Ns = 293 - 296.
*p<.05. **p<.01. ***p<.001.
ported involvement (Table 11.2). These data imply that having a very best
friendship with many positive features increased students' involvement. A
plausible explanation for this finding is that a close friendship high in
quality strengthens motives to seek mutually satisfying interactions with
the best friend and other people. Students with such friendships may be
more willing to join classmates in academic activities and more eager to
participate in class discussions. To test this hypothesis, researchers might
assess the school-related motives of students with friendships varying in
quality. The motives and goals identified by Wentzel (1989, 1991b) and
other researchers (Asher & Renshaw, 1981; Nicholls, Patashnick, &
Nolen, 1985) would be good candidates for this assessment.
Students' reports on the negative features of both their very best friend-
ships and the average of their three best friendships were related to the
changes during the year in their self-reported disruption. These data imply
that having friendships high in conflicts and rivalry increased students'
disruptive behavior. Negative interactions with friends apparently spilled
over to affect students' behavior toward other classmates and teachers.
Sullivan (1953) suggested that some children compete so often with peers
that competitive motivation becomes a prominent part of their personality.
Apparently, students in more competitive friendships acquire a habit of
competing with others in many activities. These students probably look
for chances to "put down" other classmates; they probably respond force-
fully to actions by others that seemed aimed at putting them down. Simi-
larly, students who have many conflicts with friends may try harder to de-
fend their perceived rights than to adapt to others. These students may not
seek mutually satisfying resolutions to conflicts because their friends had
rarely done so with them.
The influence of friendship quality on school adjustment should not be
exaggerated. Table 11.2 shows that the measures of friendship features ac-
counted for only a small (but significant) amount of the variance in stu-
dents' involvement and disruption. Moreover, like previous researchers,
we did not find a significant effect of friendship quality on students'
grades. Even so, evidence that friendship quality is a predictor of changes
in some aspects of school adjustment is both theoretically and practically
significant.
Age changes in the effects of friendship quality

As one aspect of positive friendship quality, intimacy first becomes an im-
portant feature of friendships in early adolescence (Berndt & Savin-
Williams, 1993; Hartup, 1992). The intimacy of friendships increases fur-
ther during adolescence, as time spent with friends increases
(Csikszentmihalyi & Larson, 1984; Sharabany, Gershoni, & Hofman,
1981). Moreover, as adolescents approach adulthood, their friendships de-
velop more positive features and become more like the ideal friendships
of classical literature (Furman & Buhrmester, 1992).
Less information is available on the negative features of friendships.
On structured tasks, adolescents sometimes compete less intensely with
friends than do elementary-school children (Berndt et al., 1986). Conflicts
with friends change little in frequency during the elementary and middle-
school years, but they may decrease during the senior-high years (Berndt
& Perry, 1986; Furman & Buhrmester, 1992; Laursen & Collins, 1994).
Changes in friendship features might be accompanied by changes in
the effects of friendship. As adolescents develop friendships higher in
quality, and friends start to interact more often, variations in friendship
quality could have a greater influence on their behavior and development.
Thus far, only one study has tested this hypothesis.
Buhrmester (1990) asked early adolescents (10- to 13-year-olds) and
middle adolescents (14- to 16-year-olds) about the intimacy of one of their
closest friendships. The adolescents also reported on their sociability, hos-
tility, and other aspects of their socioemotional adjustment. The correla-
tions of intimacy with socioemotional adjustment were stronger in middle
adolescence than in early adolescence. Although other interpretations are
possible, these findings are consistent with the hypothesis that friendship
quality has stronger effects on students' psychological adjustment as they
grow older. Furthermore, both empirical research and everyday observa-
tions show that friendships become more important to students as they
move through adolescence. Thus it is reasonable to assume that friendship
quality also becomes more important for students' adjustment.
Implications of friends9 influence for educational practice

Many teachers face the practical issue of whether to intervene in the
friendships of students who are a bad influence on each other. Should the
friends be moved to different parts of the classroom? If their new loca-
tions do not end their misbehavior, should one of them be transferred to
another class? For a full answer to these questions, we need again to con-
sider both pathways of friends' influence.
Changing the influence of friends' characteristics

Friends certainly can have a negative influence on one another (Berndt &
Keefe, in press; Epstein, 1983; Kandel, 1978). For example, friends can
discourage students' involvement in class activities. These negative influ-
ences can be reduced by separating students from their friends. Ball
(1981) observed classrooms for early adolescents before and after acade-
mic tracking was eliminated from a school. When tracking was the norm,
low-achieving students typically attended classes and formed friendships
with other low-achieving students. In these classes, students and their
friends usually had negative attitudes toward school. The classes were so
disruptive that teachers found them extremely unpleasant.
After tracking was eliminated, many disruptive and low-achieving stu-
dents found themselves in classrooms with other students average or high
in achievement. The other students typically had positive attitudes toward
school, and they did not go along with the low achievers' disruptive be-
havior. Moreover, the disruptive students did not have enough friends in
the new classrooms to form cohesive groups.
Ball's observations suggest that weakening friendships among disrup-
tive students by eliminating tracking can reduce negative influences of
friends on students' behavior. Because his study was not an experiment,
however, other explanations for the results are possible. For example, the
changes in students' behavior might be attributed to other positive effects
of eliminating tracking (see Oakes, 1985), rather than a decrease in the
negative effects of disruptive friends. The alternative explanations are not
especially relevant to our chapter, but the general issue of teachers' inter-
ference with students' friendships is relevant.
What happens to disruptive students when teachers' intervention weak-
ens their friendships? These students may find it hard to make new
friends, because students who misbehave in class are usually unpopular
with peers (Coie, 1990; Wentzel, 1991a). If the students have difficulty
making new friends, they may also have difficulty satisfying their need for
social approval. Without friends to give them praise and encouragement,
these students are likely to drop out of school, officially or unofficially
(Parker &Asher, 1987).
Instead of trying to break up friendships among students who are poor-
ly adjusted to school, teachers might try to correct students' mispercep-
tions about their friends' attitudes and behavior. Research on alcohol and
drug abuse has shown that adolescents often assume their friends have
more positive attitudes toward the use of alcohol and other drugs than they
actually do (Cook, Anson, & Walchli, 1993). Substance-abuse interven-
tions are effective partly because they reduce these misperceptions. In oth-
er words, they give adolescents accurate information about how their
friends think and act. By doing so, they reduce adolescents' motivation to
seek social approval or self-enhancement by using drugs themselves.
Similar misperceptions may exist in the realm of school adjustment.
Students may believe that their friends like school less and approve of
misbehavior more than is actually true. If teachers asked all students to re-
port their attitudes toward school, both they and their students might dis-
cover that pro-school attitudes are widely shared. Doing such a survey and
giving students the results could reduce students' misperceptions that their
friends admire classmates who are poorly adjusted to school. Reducing
these misperceptions could, in turn, enhance the positive effects of friends
on school adjustment.
Changing friendship features

When faced with friends who are a bad influence on one another, teachers
might also consider the quality of these friendships. As noted earlier, stu-
dents who are poorly adjusted to school often have friendships that are
low in quality (Berndt & Keefe, in press; Dubow et al, 1991). Instead of
trying to end these friendships or, conversely, taking a completely hands-
off attitude toward them, teachers might try to improve them and so im-
prove students' adjustment to school.
To improve students' friendships, teachers could use cooperative-learn-
ing techniques for academic instruction (Cohen, 1994; Furman & Gavin,
1989). Placing students in small groups and asking them to work together
on academic tasks has been viewed mainly as a means of raising students'
achievement (Slavin, 1983). But cooperative-learning programs also en-
hance prosocial behavior and tolerance of other people while reducing
competition between classmates (Furman & Gavin, 1989).
Cooperative-learning programs are not foolproof. Teachers need to
prepare students carefully and monitor interactions among the students in
a group. Otherwise, students with low status in the classroom may contin-
ue to be treated negatively in their small group (Cohen, 1994). Yet with
appropriate monitoring, the use of cooperative learning may not only im-
prove classroom climate but also contribute to the formation of high-qual-
ity friendships (Hansell & Slavin, 1981).
Programs for training social skills might also affect friendship quality
(Coie & Koeppl, 1990; Mize & Ladd, 1990). Students can be trained to
ask polite questions, to make suggestions, and to offer support to their
classmates. In addition, students can be trained to avoid unnecessary con-
flicts with peers and to resolve conflicts effectively when they arise. One
strategy for training emphasizes anger control, or avoiding impulsive ac-
tion when upset by a classmate's behavior. Another strategy emphasizes
social problem solving, or thinking about ways to resolve conflicts with-
out using aggression (Lochman, 1985). Because conflicts with friends
spill over to affect students' behavior toward other people (Berndt &
Keefe, in press), training in conflict resolution could reduce behavior
problems in the classroom.
One limitation in most social-skills training programs is a lack of em-
phasis on students' motivation. The programs teach students how to be-
have in social situations, but not why they should behave that way. Some
students, however, may enjoy competing more than compromising, or be
more motivated to get things they want than to develop positive relation-
ships with peers (Putallaz & Sheppard, 1990). These students need to be
persuaded that improving their friendships is worth the effort.
Trying to increase students' motivation to develop high-quality friend-
ships could be seen as risky. As students' interest in having good friend-
ships increases, their interest in strictly academic aspects of school might
be expected to decrease. Low-achieving students typically endorse the
goals of making friends and having fun at school more than do students
with high grades (Wentzel, 1989). At first glance, intervening to improve
students' friendships might be thought to enhance this contrast. The inter-
vention might encourage poorly adjusted students to work even less on
their school work than on their social life.
This concern reflects a misunderstanding of the proposed intervention.
It is designed not to increase the time students spend with friends but to
improve the quality of their friendships. Low-achieving students already
interact frequently with friends (Csikszentmihalyi & Larson, 1984). The
problem is that their friendships are often low in positive features and high
in negative features. Giving these students the motivation to be more sup-
portive to friends and to have more harmonious interactions with friends
is not likely to reduce their school adjustment. On the contrary, the change
in motivation should enhance students' social and academic competence.
Unanswered questions and future directions

Currently, evidence on the motives underlying friends' influence is largely
indirect. Few researchers have tried to assess the motives that account for
the influence of friends' characteristics. Even fewer have tried to examine
the motives that explain how and why friendship quality affects school ad-
justment.
This state of affairs is problematic. Without direct assessments of mo-
tives, it is difficult to propose convincing alternatives to the idea that stu-
dents usually conform to friends' pressure because of fear of punishment.
Without direct assessments of motives, it is difficult to explain why
friendship quality affects some aspects of school adjustment more than
others.
Two methods of assessing motives might be used in future research.
The first is an experimental approach. For many years, social psycholo-
gists have used experimental manipulations to probe the motives underly-
ing people's behavior. For example, many experiments have been done to
see whether group discussions affect people's decisions because people
are motivated to seek social approval or because they want to make cor-
rect decisions (Isenberg, 1986). Interest in these two types of motives re-
mains strong within social psychology (e.g., Wood, Lundgren, Ouellette,
Busceme, & Blackstone, 1994). Adapting the experimental manipulations
used in this research to study friends' influence on students' adjustment to
school could be very productive.
The second method of assessing students' motives is to ask them di-
rectly. As mentioned earlier, this method has been used to examine mo-
tives or goals associated with classroom behavior and academic achieve-
ment (Nicholls et al., 1985; Wentzel, 1989, 1991a, b). One limitation of
the method is that students may not be aware of the motives that govern
their behavior. Some students may also be unwilling to report their mo-
tives to a researcher. Nevertheless, these limitations should not be overes-
timated. Previous studies have shown that valuable information can be ob-
tained with direct questions about motivation. Extensions of the research
to explore the motives underlying friends' influence should also be re-
warding.
Another direction for future research would be a thorough analysis and
investigation of relationship motives. Sullivan's (1953) proposals about
friends' collaboration in pursuit of mutually satisfying outcomes were
novel in 1953, and they are still unusual. Few theories of motivation take
account of people's social relationships; even fewer include motives that
apply to partners in a relationship as a unit (see Weiner, 1992). Theories of
motivation generally deal with only two classes of motives, individualistic
(or self-centered) and altruistic (or other-centered).
The individualistic motives that affect friendship quality have some-
times been examined systematically. Students have been asked to do tasks
with a partner and then to say whether they tried to compete with the part-
ner, to get many rewards for themselves, or to pursue other goals (see
Berndt et al., 1986). Other researchers have constructed tasks that allow
the assessment of students' motives from their actual behavior rather than
from their self-reports (e.g., Knight, Dubro, & Chao, 1985). Extensions of
this work to probe the motives of students with friendships high in nega-
tive features would be worthwhile.
Motives associated with positive features of friendship, with a concern
for what "we" need instead of what "I" need, have so far not been exam-
ined directly. We have argued that these relationship motives are stronger
when friendships are more intimate, egalitarian, and have other positive
features. This argument should be evaluated more carefully.
One direction for research would be to explore the parallel between re-
lationship motives and collectivist values. Many researchers have as-
sumed that Western cultures emphasize individualistic values like person-
al freedom, whereas collectivist cultures emphasize responsibility to the
other members of an ingroup such as the family (Triandis, 1989; 1990).
People in collectivist cultures are assumed not to distinguish between their
personal goals and the goals of their small group. They are assumed to
view the success of everyone in their ingroup as their greatest goal. In this
sense, they are like close friends who seek not their own satisfaction but
mutually satisfying outcomes.
To document the contrast between individualistic and collectivist cul-
tures, researchers have often used or adapted Rokeach's (1973) survey of
values. Researchers have also devised items for assessing attitudes consis-
tent with these two orientations (see Triandis, 1990). This research could
provide not only some specific ideas but also a general approach to the
study of relationship motives. This approach could be doubly rewarding
because variations in friendships and cultural variations could be explored
simultaneously (cf. Miller, Bersoff, & Harwood, 1990).
Finally, throughout this chapter we have emphasized the distinction be-
tween the two pathways of friends' influence. This distinction may be ab-
solute. That is, the influence of friends' characteristics may be entirely in-
dependent of the effects of friendship features. Consider, for example,
students who have high-quality friendships with peers who have negative
attitudes toward school. Our argument to this point is that these students
will be positively affected by the quality of their friendships and negative-
ly affected by their friends' attitudes toward school. The net effect will de-
pend on the strength of the two separate effects.
An alternative hypothesis is that the quality of students' friendships
modifies the influence of friends' characteristics. When students trust and
admire their friends, the friends' characteristics should have an especially
powerful influence on the students' adjustment (Hallinan, 1983). Cauce
and Srebnik (1989) applied this general hypothesis to the case of school
adjustment. They argued that students' adjustment to school may worsen
greatly if they have highly supportive friendships with peers who are
poorly adjusted to school.
Cauce and Srebnik (1989) cited only correlational data in support of
their hypothesis. Findings consistent with the hypothesis were also ob-
tained in one experimental study (Berndt et al., 1990). As mentioned earli-
er, this study included pairs of friends who discussed dilemmas concern-
ing school work. The similarity of friends' decisions increased after the
discussions, showing that they influenced one another's decisions.
Friends' similarity increased most when their interactions were judged by
observers as most cooperative and least aggressive. Friends' similarity in-
creased least when they reported that their friendships were high in con-
flicts and rivalry. These results imply that friends were more influenced by
their discussions when their friendships were high in positive features and
low in negative features.
This experiment cannot be taken as conclusive because the findings
are inconsistent with those from other studies. Hundreds of studies have
suggested that supportive social relationships almost invariably have posi-
tive effects on psychological adjustment and even on physical health
(Sarason et al. 1990). Moreover, data from our longitudinal study (Berndt
& Keefe, in press) did not support the hypothesis that high friendship
quality can magnify the negative influence of poorly adjusted friends. We
looked for interactions between the effects of friends' characteristics and

of friendship features. That is, we checked to see whether the effects of
friends' characteristics varied with the quality of students' friendships, but
we found no interactions of this type.
The available data are too limited to answer questions about the rela-
tions between the two influence pathways. More attention to the contrast-
ing hypotheses is needed, because they are linked to critical assumptions
in important theories of social influence. The contrasting hypotheses also
have important practical implications. Their systematic comparison
should be a central focus of future research.
Conclusions
One message of the chapter is that friends' influence on students' adjust-
ment to school is a more complex phenomenon than most popular and
scholarly writers have implied. Interactions with friends affect students'
attitudes toward school, behavior in class, and academic achievement
through two distinct pathways. The effects of influence via each pathway
may be either to increase or to decrease students' adjustment to school.
Most importantly, influence via each pathway depends on multiple mo-
tives.
The influence of friends' characteristics has been emphasized by popu-
lar writers who express concern about the negative effects of friends' pres-
sure on students' behavior. Our review of empirical research has shown
that coercive pressure is rarely applied in friendship groups. Students are
affected by their friends' characteristics, but fear of punishment for non-
conformity to friends' pressure is not the primary motive underlying this
influence. More important are students' need for social approval, their
identification with friends, a motive for self-enhancement, and the need to
be correct or make reasonable decisions. Evidence for these four motives
is indirect, but their importance cannot be questioned.
The second pathway of influence, through the quality of students'
friendships, is linked to theories of social and personality development.
The quality of a friendship is indicated by its intimacy, by the friends'
prosocial behavior toward each other, and by the frequency of other types
of positive interactions. Negative interactions also occur between friends,
and friendship quality is lower when conflicts and rivalry are frequent and
intense. Ideally, interactions between friends are governed by a motive to
seek mutually satisfying outcomes. That is, friends think about what "we"
want, not what "I" want. In real life, friends often have motives to get their
own way or to compete with each other. Students develop more positive
attitudes toward school when their friendships are higher in quality. Stu-
dents become more disruptive at school when their friendships are lower
in quality. The connections of these effects to differences in motives have
not been documented precisely, and more research on these connections is
needed.
Information about the two influence pathways and their associated mo-
tives has clear implications for educational practice. One set of issues con-
cerns friends who negatively influence one another's adjustment to school.
Breaking up these friendships, for example, by transferring the students to
different classes, can reduce this problem, but this solution is not ideal. A
better alternative is making all students more aware of their classmates' at-
titudes toward school and beliefs about acceptable behavior. In addition,
teachers can try to improve the quality of students' friendships by using
cooperative learning or social skills training.
Our recommendations for teachers are offered tentatively, because
their basis in research is limited. Our motivational analysis derives more
from theories of social influence than from specific studies. Systematic
exploration of the motives underlying friends' influence is needed. Re-
search on possible links between the two influence pathways is also need-
ed. Future research should not only answer basic questions about how
friends influence one another during childhood and adolescence. This re-
search should also clarify how this influence can be channeled to enhance
students' adjustment to school.
References
Ansley, L., & McCleary, K. (1992, August 21-23). Do the right thing. USA Week-
end, 4-7.
Aristotle (1962). Nicomachean ethics (M. Ostwald, trans.). New York: Bobbs-
Merrill, 1962.
Asher, S. R., & Renshaw, P. D. (1981). Children without friends: Social knowl-
edge and social skill training. In S. R. Asher & J. M. Gottman (Eds.), The de-
velopment of children s friendships (pp. 273-296). Cambridge, England: Cam-
Ball, S. J. (1981). Beachside comprehensive. Cambridge, England: Cambridge
Press.
Bandura, A. (1969). Social learning theory of identificatory processes. In D. A.
Goslin (Ed.), Handbook of socialization theory and research (pp. 213-262).
Chicago: Rand-McNally.
Bandura, A. (1977). Social learning theory. Englewood Cliffs, NJ: Prentice Hall.
Bandura, A., & Walters, R. (1963). Social learning and personality development.
New York: Holt, Rinehart and Winston.
Berndt, T. J. (1986). Sharing between friends: Contexts and consequences. In E.
C. Mueller and C. Cooper (Eds.), Process and outcome in peer relationships
(pp. 129-160). New York: Academic.
Berndt, T. J. (1992). Friendship and friends' influence in adolescence. Current Di-
rections in Psychological Science, 1, 156-159.
Berndt. T. J., Hawkins, J. A., & Hoyle, S. G. (1986). Changes in friendship during
a school year: Effects on children's and adolescents' impressions of friendship
and sharing with friends. Child Development, 57, 1284-1297.
Berndt, T. J. & Keefe, K. (1995). Friends' influence on adolescents' adjustment to
school. Child Development, 66, 1312-1329.
Berndt, T. J., & Laychak, A. E., & Park, K. (1990). Friends' influence on adoles-
Berndt, T. J., McCartney, K. A., Caparulo, B. K., & Moore, A. M. (1983-1984).
The effects of group discussions on children's moral decisions. Social Cogni-
tion, 2, 343-360.
Berndt, T. J., & Miller, K. A. (1990). Expectancies, values, and achievement in ju-
nior high school. Journal of Educational Psychology, 82, 319-326.
Berndt, T. J., Miller, K. A., & Park, K. (1989). Adolescents' perceptions of
friends' and parents' influence on aspects of school adjustment. Journal ofEar-
ly Adolescence, 9, 419-435.
Berndt, T. J. & Perry, T. B. (1986). Children's perceptions of friendships as sup-
portive relationships. Developmental Psychology, 22, 640-648.
Berndt, T. J., & Savin-Williams, R. C. (1993). Variations in friendships and peer-
group relationships in adolescence. In P. Tolan & B. Cohler (Eds.), Handbook
of clinical research and practice with adolescents (pp. 203-219). New York:
Wiley.
Bishop, J. H. (1989). Why the apathy in American high schools? Educational re-
searcher, 7S(1), 6-10.
Brown, B. B., Clasen, D. R., & Eicher, S. A. (1986). Perceptions of peer pressure,
peer conformity dispositions, and self-reported behavior among adolescents.
Buhrmester, D. (1990). Intimacy of friendship, interpersonal competence, and ad-
justment during preadolescence and adolescence. Child Development, 61,
1101-1111.
Buhrmester, D., & Furman, W. (1986). The changing functions of friends in child-
hood: A neo-Sullivanian perspective. In V J. Derlega & B. A. Winstead (Eds.),
Friendship and social interaction (pp. 41-62). New York: Springer-Verlag.
Bukowski, W. M., Hoza, B., & Boivin, M. (1994). Measuring friendship quality
during pre- and early adolescence: The development and psychometric proper-
ties of the Friendship Qualities Scale. Journal of Social and Personal Relation-
ships,17,47r 1-484.
Cauce, A. M., Srebnik, D. S. (1989). Peer networks and social support: A focus
for preventive effects with youth. In L. A. Bond & B. E. Compas (Eds.), Prima-
ry prevention and promotion in the schools (pp. 235-254). Newbury Park, CA:
Sage.
Chaiken, S., Liberman, A., & Eagly, A. H. (1989). Heuristic and systematic pro-
cessing within and beyond the persuasion context. In J. S. Uleman & J. A.
Bargh (Eds.), Unintended thought (pp. 212-252). New York: Guilford.
Cohen, E. G. (1994). Restructuring the classroom: Conditions for productive
small groups. Review ofEducational Research, 64, 1-35.
Coie, J. (1990). Toward a theory of peer rejection. In S. R. Asher and J. D. Coie
(Eds.), Peer rejection in childhood (pp. 365-402). Cambridge, England: Cam-
Coie, J. D., & Koeppl, G. K. (1990). Adapting intervention to the problems of ag-
gressive and disruptive rejected children. In S. R. Asher and J. D. Coie (Eds.),
Peer rejection in childhood (pp. 309-337). Cambridge, England: Cambridge
University Press.
Coleman, J. S. (1961). The adolescent society. New York: Free Press.
Cook, T. D., Anson, A. R., & Walchli, S. B. (1993). From causal description to
causal explanation: Improving three already good evaluations of adolescent
health program. In S. G. Millstein, A. C. Petersen, & E. O. Nightingale (Eds.),
Promoting the health of adolescents (pp. 339-374). New York: Oxford Univer-
sity Press.
Crandall, V J., Katkovsky, W., & Preston, A. (1960). A conceptual formulation for
some research on children's achievement development. Child Development, 31,
787-797.
Csikszentmihalyi, M., & Larson, R. (1984). Being adolescent. New York: Basic
Books.
Damon, W., & Phelps, E. (1989). Strategic uses of peer learning in children's edu-
cation. In T. J. Berndt & G. W. Ladd (Eds.), Peer relationships in child develop-
ment (pp. 135-157). New York: Wiley.
Davies, M., & Kandel, D. B. (1981). Parental and peer influences on adolescents'
educational plans: Some further evidence. American Journal of Sociology, 87,
363-387.
Deutsch, M., & Gerard, H. B. (1955). A study of normative and informational so-
cial influences upon individual judgment. Journal of Abnormal and Social Psy-
chology, 51, 629-636.
Downs, W. R. (1987). A panel study of normative structure, adolescent alcohol
use, and peer alcohol use. Journal of Studies on Alcohol, 48, 167-175.
DuBois, D. L., Felner, R. D., Brand, S., Adan, A. M, & Evans, E. G. (1992). A
prospective study of life stress, social support, and adaptation in early adoles-
cence. Child Development, 63, 542-557.
Dubow, E. G., Tisak, J., Causey, D., Hryshko, A., & Reid, G. (1991). A two-year
longitudinal study of stressful life events, social support, and social problem-
solving skills: Contributions to children's behavioral and academic adjustment.
Epstein, J. L. (1983). The influence of friends on achievement and affective out-
comes. In J. L. Epstein & N. Karweit (Eds.), Friends in school: Patterns of se-
lection and influence in secondary schools (pp. 177-200). New York: Academ-
ic Press.
Furman, W., & Buhrmester, D. (1992). Age and sex differences in perceptions of
networks of personal relationships. Child Development, 63, 103-115.
Furman, W., & Gavin, L. A. (1989). Peers' influence and adjustment and develop-
ment: A view from the intervention literature. In T. J. Berndt & G. W. Ladd
(Eds.), Peer relationships in child development (pp. 319-340). New York: Wi-
ley.
Hallinan, M. T. (1978/1979). The process of friendship formation. Social Net-
works, 1, 193-210.
Hallinan, M. T. (1983). Commentary: New directions for research on peer influ-
ence. In J. L. Epstein & N. Karweit (Eds.), Friends in school: Patterns of selec-
tion and influence in secondary schools (pp. 219-231). New York: Academic
Press.
Hansell, S., & Slavin, R. E. (1981). Cooperative learning and the structure of in-
terracial friendships. Sociology of Education, 54, 98-106.
Hartup, W. W. (1983). Peer relations. In P. H. Mussen (Series Ed.), E. M. Hether-
ington (Vol. Ed.), Handbook of child psychology. Vol. 4. Socialization, person-
ality, and social development (pp. 103-196). New York: Wiley.
Hartup, W. W. (1992). Friendships and their developmental significance. In H.
McGurk (Ed.), Contemporary issues in childhood social development. Hove,
England: Erlbaum.
Hoving, K. L., Hamm, N., & Galvin, P. (1969). Social influence as a function of
stimulus ambiguity at three age levels. Developmental Psychology, 1, 631-
636.
Isenberg, D. J. (1986). Group polarization: A critical review and meta-analysis.
Journal of Personality and Social Psychology, 50, 1141-1151.
Juvonen, J., & Murdock, T. B. (1993). How to promote social approval: Effects of
audience and achievement outcome on publicly communicated attributions.
Juvonen, J., & Weiner, B. (1993). An attributional analysis of students' interac-
tions: The social consequences of perceived responsibility. Educational Psy-
chology Review, 5, 325-345.
Kandel, D. B. (1978). Homophily, selection, and socialization in adolescent
Kandel, D. B., & Andrews, K. (1987). Processes of adolescent socialization by
parents and peers. International Journal of the Addictions, 22, 319-342.
Keefe, K. (1994). Perceptions of normative social pressure and attitudes toward
alcohol use: Changes during adolescence. Journal of Studies on Alcohol', 55,

46-54.
Kelman, H. C. (1961). Processes of opinion change. Public Opinion Quarterly,
25, 57-78.
Knight, G. P., Dubro, A. E, & Chao, C.-C. (1985). Information processing and the
development of cooperative, competitive, and individualistic social values. De-
velopmental Psychology, 21, 37^5.
Kurdek, L. A., & Sinclair, R. J. (1988). Adjustment of young adolescents in two-
parent nuclear, stepfather, and mother-custody families. Journal of Consulting
and Clinical Psychology, 56, 91-96.
Laursen, B. (1993). Conflict management among close peers. In B. Laursen (Ed.),
Close friendships in adolescence (pp. 39-54). San Francisco, CA: Jossey-Bass.
Laursen, B., & Collins, W. A. (1994). Interpersonal conflict during adolescence.
Psychological Bulletin, 115, 197-209.
Lochman, J. E. (1985). Effects of different treatment lengths in cognitive behav-
ioral interventions with aggressive boys. Child Psychiatry and Human Devel-
opment, 16, 45—56.
Maccoby, E. E., & Martin, J. A. (1983). Socialization in the context of the family:
Parent-child interaction. In E. M. Hetherington (Ed.), Socialization, personali-
ty, and social development. Vol. 4. Handbook of child psychology (pp. 1-101).
New York: Wiley.
Marsh, H. W. (1989). Age and sex effects in multiple dimensions of self-concept:
Preadolescence to early adulthood. Journal of Educational Psychology, 81,
417^30.
Miller, J. G., Bersoff, D. M., & Harwood, R. L. (1990). Perceptions of social re-
sponsibilities in India and in the United States: Moral imperatives or personal
decisions? Journal of Personality and Social Psychology, 58, 33^7.
Mischel, W. (1970). Sex typing and socialization. In P. H. Mussen (Ed.),
CarmichaeVs Manual of Child Psychology. Vol. II (pp. 3-72). New York: Wiley.
Mize, J., & Ladd, G. W. (1990). Toward the development of successful social skills
training for preschool children. In S. R. Asher and J. D. Coie (Eds.), Peer rejec-
tion in childhood (pp. 338-361). Cambridge, England: Cambridge University
Press.
Nicholls, J. G., Patashnick, M., & Nolen, S. B. (1985). Adolescents' theories of
education. Journal of Educational Psychology, 77, 683-692.
Nisbett, R. E., & Wilson, T. D. (1977). Telling more than we can know: Verbal re-
ports on mental processes. Psychological Review, 84, 231-259.
Oakes, J. (1985). Keeping track. New Haven: Yale University Press.
Are low-accepted children "at risk"? Psychological Bulletin, 102, 357-389.
Parker, J. G., & Asher, S. R. (1993). Friendship and friendship quality in middle
childhood: Links with peer group acceptance and loneliness. Developmental
Piaget, J. (1965). The moral judgment of the child. New York: The Free Press.
(Originally published, 1932.)
Putallaz, M., & Sheppard, B. H. (1990). Social status and children's orientations
to limited resources. Child Development, 61, 2022-2027.
Rokeach, M. (1973). The nature of human values. New York: Free Press.
Ruble, D. N., Boggiano, A. K., Feldman, N. S., & Loebl, J. H. (1980). Develop-
mental analysis of the role of social comparison in self-evaluation. Develop-
Sarason, B. R., Sarason, I. G., & Pierce, G. R. (Eds.) (1990). Social support: An
interactional view. New York: Wiley.
Savin-Williams, R. C , & Berndt, T. J. (1990). Friendships and peer relations
during adolescence. In S. S. Feldman & G. Elliott (Eds.), At the threshold: The
developing adolescent (pp. 277-307). Cambridge, MA: Harvard University
Press.
Schunk, D. H., Hanson, A. R., & Cox, P. D. (1987). Peer-model attributes and
children's achievement behaviors. Journal of Educational Psychology, 79, 5 4 -
61.
Schwartz, F. (1981). Supporting or subverting learning: Peer group patterns in
four tracked schools. Anthropology and Education Quarterly, 13, 99-121.
Sebald, H. (1986). Adolescents' shifting orientations toward parents and peers: A
curvilinear trend over recent decades. Journal of Marriage and the Family, 48,
5-13.
Sharabany, R., Gershoni, R., & Hofman, J. E. (1981). Girlfriend, boyfriend: Age
and sex differences in intimate friendship. Developmental Psychology, 17,
800-808.
Slavin, R. E. (1983). When does cooperative learning increase student achieve-
ment. Psychological Bulletin, 94, 429-445.
Smith, H. W. (1973). Some developmental interpersonal dynamics through child-
hood. American Sociological Review, 38, 543—552.
Steinberg, L., & Silverberg, S. R. (1986). The vicissitudes of autonomy in early
Sullivan, H. S. (1953). The interpersonal theory of psychiatry. New York: Norton.
Suttles, G. D. (1968). The social order of the slum. Chicago: University of Chica-
go Press.
Tesser, A. (1984). Self-evaluation maintenance processes: Implications for rela-
tionships and for development. In J. C. Masters & K. Yarkin-Levin (Eds.),
Boundary areas in social and developmental psychology (pp. 271—299). New
chology, 46, 561-574.
Triandis, H. C. (1989). The self and social behavior in differing cultural contexts.
Psychological Review, 96, 506—520.
Triandis, H. C. (1990). Cross-cultural studies of individualism and collectivism.

Nebraska Symposium on Motivation, 57, 41-133.
Tudge, J. R. H. (1992). Processes and consequences of peer collaboration: A Vy-
gotskian analysis. Child Development, 63, 1364-1379.
Urberg, K. A., Cheng, C.-H., & Shyu, S.-J. (1991). Grade changes in peer influ-
ence on adolescent cigarette smoking: A comparison of two measures. Addic-
tive Behavior, 16,21-2$.
tion. In C. P. Smith (Ed.), Achievement-related motives in children (pp.
46-101). New York: Russell Sage Foundation.
Weiner, B. (1992). Human motivation: Metaphors, theories, and research. New-
bury Park, CA: Sage.
tional Psychology, 81, 131-142.
Wentzel, K. R. (1991a). Relations between social competence and academic
Wentzel, K. R. (1991b). Social competence at school: Relation between social re-
1-24.
357-364.
Wood, W, Lundgren, S., Ouellette, J. A., Busceme, S., & Blackstone, T. (1994).
Minority influence: A meta-analytic review of social influence processes. Psy-
chological Bulletin, 115, 323-345.
Youniss, J., & Smollar, J. (1985). Adolescent relations with mothers, fathers, and
friends. Chicago: University of Chicago Press.
Zanna, M. P., Olson, J. M., & Herman, C. P. (Eds.). (1987). Social influence.
Zimbardo, P. G., & Leippe, M. R. (1991). The psychology of attitude change and
social influence. Philadelphia: Temple University Press.
12 Peer networks and students' classroom
engagement during childhood
and adolescence
Thomas A. Kindermann, Tanya L McCollam,

and Ellsworth Gibson, Jr.
If one asks parents and teachers about important influences on children's

motivation and adjustment to school, answers will likely suggest four sets
of factors: the teacher and the general school environment, the psycholog-
ical make-up of the individual child himself or herself, the family environ-
ment, and the child's relationships with his or her peers in school. In fact,
research on school motivation and adjustment has examined all four influ-
ences. However, if one looks at current discussions of motivation and
school adjustment (e.g., Ames & Ames, 1984, 1985), most research seems
to concentrate on the first two factors, namely, the school and the child;
some efforts target the family, and only comparatively few include chil-
dren's peers.
Characteristics of schools, classrooms, teachers, and students have
been prime targets of motivational studies (Skinner & Belmont, 1993). In
general, it is educational researchers who have focused on school and
classroom contexts (for reviews, see Ames & Ames, 1985; Brophy, 1983;
1986), such as the role of teacher behaviors, teaching styles, or evaluation
strategies (Boggiano & Katz, 1991; Brophy, 1985, 1986; Graham & Bark-
er, 1990; Grolnick & Ryan, 1987; Keller, 1983; Midgley, Feldlaufer, &
Eccles, 1989; 1990; Moely et al., 1992), and the overall classroom envi-
ronment and organization (Ames, 1984; Eccles, Midgley, & Adler, 1984;
Johnson & Johnson, 1985).
Psychological research has focused more on children themselves (for
reviews see Ames & Ames, 1984; Dweck & Elliott, 1983; Stipek, 1993),
specifically on their understanding and explanations of their own role in
the school environment. Key constructs are children's attributions (Wein-
er, 1979, 1985, 1986), their beliefs about themselves and the extent to
which they feel in control (Chapman, Skinner, & Bakes, 1990; Patrick,
Skinner, & Connell, 1993; Skinner, Wellborn & Connell, 1990; Weisz &
Cameron, 1985), and their self-efficacy in the school environment
279
(Schunk, 1991). Often, this is combined with examinations of children's

perceived abilities or competencies (Mclver, Stipek, & Daniels, 1991),
goal orientations (Elliott & Dweck, 1988; Meece & Holt, 1993; Nicholls,
1984; Wentzel, 1989), learning strategies (Ainley, 1993; Pintrich & De
Groot, 1990), and interests (Schiefele, 1991), or of children's self-concept
or self-worth (Covington, 1984; Wigfield & Karpathian, 1991). Last but
not least, there is increasing interest in children's intrinsic motivation for
academic activities and their sense of autonomy within the school envi-
ronment (Connell & Wellborn, 1991; Corno & Rohrkemper, 1985; Deci &
Ryan, 1985).
Compared to both of these areas, research on other context influences
on student motivation and adjustment is relatively sparse. Family oriented
researchers have examined childrearing practices (e.g., DeBaryshe, Pat-
terson, & Capaldi, 1993; Dishion, 1990), parents' provision of autonomy
support and involvement (Connell, Spencer, & Aber, 1994; Gottfried,
Fleming, & Gottfried, 1994; Grolnick, Ryan, & Deci, 1991), parental val-
ues, expectations, and standards (Stevenson et al., 1990), as well as par-
ents' explanations for children's success and failure (Holloway & Hess,
1982).
However, even this relatively small body of empirical work is large
compared to the amount of research dedicated to the study of peer influ-
ences on school motivation and adjustment. Although the study of peer re-
lationships has a long-standing theoretical and empirical tradition among
developmentalists who are interested in social development (for reviews
see Asher & Coie, 1990; Hartup, 1978; 1983), most attention has been di-
rected toward variables like social adjustment, social behavior, and, if re-
lated to the academic domain, to achievement outcomes.
The fact that the current volume exists can be taken as evidence that
this is changing, and that social relationships are accorded a more impor-
tant role by many as a factor relevant for motivational development and
adjustment to school. In this chapter, we hope to contribute a specific per-
spective. Our chapter aims to examine the role of children's and adoles-
cents' affiliations with peer groups in their developing school motivation
during elementary and high school. After outlining our general frame-
work, which is based on a contextual understanding of development and
oriented toward a self-system view of motivation, we will describe meth-
ods that aim to identify students' peer contexts and show promise for ex-
amining their influences for students' motivation in school. To illustrate
this perspective and these methods, we will use data from two studies, one
focusing on 4th and 5th grade children (Kindermann, 1993), and a new
study on adolescents. Specific attention will be paid to two motivationally
Peers and school motivation 281
relevant processes, namely processes of how children and adolescents se-

lect other students as peer contexts for themselves, and processes of how
these peer contexts, in turn, can influence the students' own subsequent
motivational development.
Peer influences on motivation and achievement

There are many indications suggesting that children's peer relationships
can have an important role in their school adjustment and motivation, in
addition to students' own psychological profiles and their interactions
with teachers. From a motivational standpoint, theoretical expectations
exist that children's need for belongingness (Weiner, 1990), their connect-
edness to a "community of learners" (Skinner & Belmont, 1993), and
feelings of relatedness to others in the classroom (Connell & Wellborn,
1991) do extend to peers as well as adults.
Thus, many of the characteristics of students' peer relations that are as-
sumed to be influential for their social development can also be regarded
as influential for their motivational development. Particularly prominent
is the sociometric tradition, which focuses on student's overall standing
and popularity in the classroom (e.g., Coie, Dodge, & Copotelli, 1982).
Research has shown that a child's sociometric standing among his or her
classmates is a strong predictor of his or her further development (for re-
views, see Newcomb, Bukowski, & Pattee, 1993; Parker & Asher, 1987;
Price & Dodge, 1989). Of specific interest for this chapter are findings
that point to a relation between low social status and variables like risks
for school dropout (cf. Parker & Asher, 1987) and academic failure
(Green, Forehand, Beck, & Vosk, 1980).
A second tradition is the study of children's and adolescents' friend-
ships (e.g., Berndt, 1989; Berndt, Laychak, & Park, 1990; Cohen, 1977;
Hallinan & Williams, 1990; Ladd 1990; Kandel, 1978a; b; Tesser, Camp-
bell, & Smith, 1984). Friendship researchers emphasize that adjustment
does not depend only on how well one is liked or accepted overall, but
also on more reciprocal aspects of peer relationships. Hence, adjustment
depends on the individual as much as on the specific others with whom
the individual becomes affiliated. Friendship researchers argue that differ-
ent kinds of children have different kinds of friends and that friends have
important functions for children's adjustment to school. For example, at
early ages, the quantity and quality of children's friendships was found to
be a predictor of their adjustment to school (Ladd, 1990). At older ages,
friends seem to become even more important. From childhood to adoles-
cence, there is some increase in the amount of time that individuals spend
with their friends (Larson & Richards, 1991), an increase in the emotional
quality during interactions with one's friends (Csikszentmihalyi & Larson,
1984), and an increase in the extent to which the quality of one's close
friendships is related to social adjustment (Buhrmester, 1990).
One line of friendship research that is of special interest for the present
discussion is represented by studies that examine which kinds of children
or adolescents become friends with one another, and the specific process-
es of influence that occur between them. For example, Hallinan and
Williams (1990) examined (about 1400) reciprocal friendships of adoles-
cents for their effects on college aspirations and actual college attendance.
As was found in many other studies, reciprocal friends were highly similar
to one another, and the extent of similarity that existed among friends was
related to their academic behavior. Friends who were most similar with re-
gard to gender, racial characteristics, and academic tracks also had very
similar college aspirations. Interestingly, adolescents who had friends
across gender, race, or tracking barriers tended to have higher college as-
pirations than others.
Our perspective is most closely related to a third tradition, namely, the
study of children's peer group networks. Students' close friends may be
only part of the picture. Teachers and parents often believe that it is not
just children's best friends who exert a powerful influence on their adjust-
ment in school, but the larger group of peers with whom they affiliate. Of
interest are processes of influence that may exist within networks of peers
(e.g., Cairns, Cairns, & Neckerman, 1989; Feiring & Lewis, 1989; Fur-
man, 1989), and specifically, processes of peer selection (Who becomes a
member of a peer group?) and socialization (Do a child's peer group
members have an influence on that child?). Like friendship researchers,
proponents of a network approach have also investigated school-related
behavior in their studies. For example, Cairns and colleagues (1989)
found that 7th grade students who later dropped out of school were likely
to be members of peer networks that consisted of students who were also
at risk for drop out.
Defining school motivation: The concept of engagement

Our studies were guided by a conceptualization of school motivation in
terms of self-system processes (e.g., Connell, 1990; Connell & Wellborn,
1991; Skinner et al., 1990). According to this model, motivational
processes within the child are neither solely products of the child's own
characteristics, nor of his or her context, but outcomes of dialectical rela-
tions between children's psychological needs and their experiences in in-

teractions with their environment.
Engagement in the classroom is seen as the prime indicator of school
motivation, and as the outcome of the extent to which children's needs are
met by environmental characteristics at school (Cornell & Wellborn,
1991). Typically, engaged children are described as selecting tasks at the
border of their competencies, taking initiative when there is an opportuni-
ty, exerting effort and concentration when working on tasks, and persist-
ing when tasks demand more than routine effort. On the opposite pole,
children whose needs are not met by their school environment are likely to
become disaffected. These children are passive, do not try hard, and give
up easily when faced with tasks that demand more than routine exertion
(Wellborn, 1991).
Students' engagement in classroom activities has considerable long-
term consequences. On the one hand, the extent to which students are en-
gaged in ongoing learning activities sets the stage for their academic
achievement and adjustment; for example, engagement was shown to be
related to perceptions of control in the school environment as well as to
children's sense of autonomy in the classroom (Patrick et al., 1993; Skin-
ner et al., 1990). On the other hand, students' behavior is also likely to in-
fluence their social interactions with teachers as well as with other stu-
dents in the classroom. For example, students' classroom behavior can
influence teachers' opinions about their competencies (Bennett, Gottes-
man, Rock, & Cerullo, 1993), as well as teachers' expectations of their
further success (e.g., Brophy, 1983; Jussim, 1989). Across time, engaged
students are more likely to experience support from their teachers, and
tend to become even more engaged, while disaffected students are likely
to experience interactions with teachers negatively and to further decrease
in motivation (Skinner & Belmont, 1993).
Challenges of studying peer network

contexts and motivation
If there is reason to believe that peer contexts are important influences for
students' motivation in school, why have these contexts not been studied
more? Peer contexts have three properties that make them quite distinct
from any other context studied traditionally.
First, they are self-selected to a large extent. Teachers and parents are
contexts that are assigned to an individual child, whereas peer contexts
usually are not. Within the constraints of a given setting, children are rela-
tively free to affiliate with others according to their own needs and de-
sires. However, these constraints may change from childhood to adoles-
cence. In adolescence, for example, the age composition in mixed-grade
classrooms may present constraints on peer selection processes within
these classrooms, while at the same time a larger range of choices may ex-
ist, because students move across different classrooms during the school
day.
Second, peer contexts consist of multiple and overlapping groups of in-
dividuals. Students need to be regarded both as individuals and as contexts
for the other individuals with whom they share networks. In comparison
to the contexts traditionally studied as influential for student motivation,
these characteristics make peer context hard to identify.
Third, although teachers typically remain stable contexts for a school
year, and one's parents for a lifetime, children's peer affiliations may
change quite rapidly and unexpectedly. This change may occur in (at
least) two ways. Peer affiliations may change in terms of who does or does
not belong to a group. In addition, the members of a child's or adolescent's
peer groups are other children or adolescents, and we need to assume that
they themselves change and develop at the same rate as our target individ-
uals under study.
Identifying peer group networks

Researchers in the area of peer relationships are generally well aware of
these problems. The methods for identifying categories of rejected, popu-
lar, or neglected children within a classroom have been the topic of con-
siderable discussion (cf. Asher & Hymel, 1981; Hymel & Rubin, 1985;
Newcomb et al., 1993). So have the criteria for identification of social
networks (e.g., Cairns, Gariepy, & Kindermann, 1990; Wellman &
Berkowitz, 1988; Wasserman & Galaskiewicz, 1994). Perhaps the only
area within this literature that can be relatively certain about definitional
criteria is the research on children's friendships (e.g., Ladd, 1990). In
friendship research, the phenomenon is usually restricted to friendships on
which two children agree reciprocally. However, as soon as we leave the
age of childhood, the definition of a friendship usually can not be left any
more to the subjects themselves, and definitional problems are recognized
(cf. Adams, 1989).
With regard to peer groups, it is an even more critical issue to identify
who, among a group of candidates, can be assumed to be important for a
given individual. While social network researchers usually also rely on the
traditional strategy of obtaining children's reports of their own affiliations
with others, we made use of a new assessment strategy. Cairns, Perrin, and
Cairns (1985; see also Cairns, Cairns, Neckerman, Gest, & Gariepy, 1988;
Cairns et al., 1989) have developed a method for assessing children's peer
networks among each other that employs children as expert observers of
their whole classroom. Children are probably better informed about what
peers are doing than most adults (or researchers), and the goal of this
method is to assess what is publicly known about existing peer associa-
tions. Students are asked to report about "who hangs around with whom?"
in a classroom. These reports are usually based on free recall; groups of
any size can be reported, and students can be nominated as belonging to
any number of groups at the same time. In response to the probe, students
typically generate lists of names of students who belong to groups. For ex-
ample one informant may recall students ALI, BEV, and CAR to be in one
group, whereas another adds a second group consisting of GIL, HAL, and
FIN. With children, we used the original interview procedure; with ado-
lescents, we used a paper-and-pencil adaptation that was administered in a
group format.
Two specific advantages of the method should be noted before we go
into the details of its use. One advantage is that informants do not just re-
port about themselves, but about all the social configurations that are
known in the classroom. This allows us to examine peer groups in class-
rooms in which not all students participate as informants themselves; usu-
ally, participation rates of at least 50% are recommended (Cairns et al.,
1985). A second advantage is that it becomes possible to assess the consis-
tency of the individual reports with the extracted map of publicly known
group affiliations in the setting.
Analyzing group nominations

In most cases, the reports will be too complex for just a qualitative analy-
sis, and researchers may want to use reliability criteria for decisions about
students' network memberships. Identification procedures usually pro-
ceed in two steps. First, a matrix is formed of co-nominations among stu-
dents. This is a matrix of conditional nomination frequencies, given that a
specific student is nominated to have a group himself or herself (see also
Breiger's, 1988, P-matrix of "person-to-person relations"). Table 12.1
gives an example of conominations in a classroom that consisted of ado-
lescents from 9th through 1 lth grade.
The second step is to identify students' actual affiliates. Many statisti-
cal tools can be used to identify patterns in conomination matrices. Usual-
ly, these have the goal to identify overall network structures and to obtain
Table 12.1 Matrix of conominations for groups in a classroom attended by 9th through 1 lth graders.
ALI BEV CAR DAR EVE FIN GIL HAL INA JIL KEN LES MIK NIC OLA PAT QIA RIA SAL TIA ULA Total Nominations
ALI — 18 19 19 18 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 20
BEV 18 — 17 17 16 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 18
CAR 19 17 — 20 17 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 20
DAR 19 17 20 — 17 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 20
EVE 18 16 17 17 — 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 18
FIN 0 0 0 0 0 — 20 20 5 3 3 3 2 0 0 0 0 0 0 0 0 29
GIL 0 0 0 0 0 20 — 20 5 4 4 4 3 0 0 0 0 0 0 0 0 21
HAL 0 0 0 0 0 20 20 — 5 3 3 3 2 0 0 0 0 0 0 0 0 20
INA 0 0 0 0 0 5 5 5 — 7 14 5 7 0 0 0 0 0 0 0 0 17
JIL 0 0 0 0 0 3 4 3 7 — 11 14 15 0 0 0 0 0 0 0 0 16
KEN 0 0 0 0 0 3 4 3 14 11 — 9 11 0 0 0 0 0 0 0 0 33
LES 0 0 0 0 0 3 4 3 5 14 9 — 13 0 0 0 0 0 0 0 0 14
MIK 0 0 0 0 0 2 3 2 7 15 11 13 — 0 0 0 0 0 0 0 ; 16
NIC 0 0 0 0 0 0 0 0 0 0 0 0 0 — 8 8 8 2 2 2 1 11
OLA 0 0 0 0 0 0 0 0 0 0 0 0 0 8 — 9 16 5 4 2 2 18
PAT 0 0 0 0 0 0 0 0 0 0 0 0 0 8 9 — 9 2 2 1 1 12
QIA 0 0 0 0 0 0 0 0 0 0 0 0 0 8 16 9 — 5 2 0 0 16
RIA 0 0 0 0 0 0 0 0 0 0 0 0 0 2 5 2 5 — 9 3 2 12
SAL 0 0 0 0 0 0 0 0 0 0 0 0 0 2 4 2 2 9 — 5 4 11
TIA 0 0 0 0 0 0 0 0 0 0 0 0 0 2 2 1 0 3 5 — 11 13
ULA 0 0 0 0 0 0 0 0 0 0 0 0 0 1 2 1 0 2 4 — 11
Matrix shows the number of times each student in the classroom was reported to be in a group together with any other student. A considerable number of conominations with stu-
dents from other classrooms is omitted. Across all classrooms, 323 groups were generated by 73 informants; the total of group nominations was 1082. Six students were not nomi-
nated to be in any group in this classroom.
a structural description of an entire setting (see, for example, Cairns et al.,

1990; Wellman & Berkowitz, 1988). However, we approached the ques-
tion from a different angle, namely, from the perspective of individuals.
Our question was not about network structures in general, but about who,
among many candidates, can be considered to be a relevant context for a
target child, and who cannot. Thus, there was little need to describe the
overall environment, as is usually attempted by sociologists and social
network researchers. Instead, the goal is to identify the exact peers with
whom a student is affiliated.
Borrowing from strategies for analyzing social interactions (e.g., Bake-
man & Gottmann, 1986; Sackett, Holm, Crowley, & Henkins, 1979), we
focused on conditional probabilities in patterns of conominations: Given
that a specific individual has been nominated to be in a group with other
students, how likely is it that any other individual is nominated to belong
to the same group? And, are the conditional probabilities for these other
individuals higher than could be expected by chance? Chance expecta-
tions are based on the nominations that the candidates had received
among all group nominations. Discrepancies between observed (condi-
tional) probabilities and their (expected) base rates can be tested via bino-
mial z-tests (or using Fisher's exact test in cases of low expected frequen-
cies; for ease of computation we use Stirling's approximation formula; see
von Eye, 1990).
To give an example (see also Table 12.1) in a classroom of 9th through
1 lth grade adolescents, student ALI was nominated 20 times, and student
BEV was nominated 18 times to be a member of a group, out of a total of
323 groups that were generated by the respondents. Thus, the expected
rate for BEV to be nominated in any group of students was .06 (18/323).
However, given that ALI was nominated to belong to a group, BEV was
nominated to be a member of the same group in 18 of the 20 cases, yield-
ing a conditional probability of .90; the z-score of 16.99 is highly signifi-
cant, denoting that ALI and BEV are members of the same group. These
tests were conducted for all combinations of students in the classrooms
under study. Connections that were found to be significant at the 5% level
are depicted in Figure 12.1.
It should be noted that in this figure, no distinction is made with regard
to individuals' centrality in the whole setting or within their networks; in-
dividuals' positions are arbitrary and based on drawing convenience only.
The method should mainly be considered as a tool to partition social
ecologies into different kinds of contexts, namely, a context that can be
considered to be central for a specific individual (one's own network of
RON
' Six connections to

TIM
| male and female 10th
(10) i and 11th graders in
1
other classroom
ZAC
(11)
Two connections
to male 10th
graders in other
classroom
Ten connections
to male and female
9th and 10th graders
in other classroom
are
Note: D a r e male. O female students.
Students' grade levels are given in parentheses.
Figure 12.1. Map of social networks in a science classroom attended by

9th through 1 lth graders (p < .05)
"buddies") and another context that is assumed to be of minor relevance

(one's other classmates, or "bystanders").
Peer group profiles as descriptors of groups 'motivational

characteristics
Classical accounts (e.g., Moreno, 1934) as well as current sociological
strategies of network analysis (see Wasserman & Galaskiewicz, 1994;
Wellman & Berkowitz, 1988) usually focus on structural characteristics to
describe peer networks, as, for example, group size, cohesiveness, indi-
viduals' centrality within groups, or groups' centrality within the larger
setting. However, we were more interested in the psychological character-
istics of peer groups than in their structural parameters.
An assumption that became central for our efforts to describe peer af-
filiations was that groups can be described as a joint function of the char-
acteristics of their members. This may be debatable, because it leads to
aggregation of scores across all of the individual members of a given stu-
dent's peer group. However, aggregation has several advantages. One ad-
vantage is the possibility of forming a group score or profile in order to
express group characteristics in one variable. A second advantage is that
this group score then allows us to compare networks that differ in terms of
structural characteristics (size, overlapping memberships, etc.). Third,
across time, group change is likely to involve turnover in group member-
ships, and we may not want to limit ourselves to examining influences
only within stable networks.
A simple strategy for capturing the composite profile of a child's peer
group(s) is to average the scores of the members of his or her network. In
the example in Figure 12.1, this means that ALI's peer group score is the
average of the scores of BEV, CAR, DAR, and EVE. All members of her
peer group are considered to be of equal importance. However, alternative
strategies are also possible; individual weights can be used in the averag-
ing procedure if there is reason to assume that some individuals carry
more weight than others, or standard deviations can be used if target ques-
tions are directed at examining group diversity.
Examining peer group processes

What we have discussed so far is a way to identify a student's networks of
peers at some level of reliability, and a way to form a representation for
the characteristics of this entire peer group. Taken together, these methods
offer strategies to examine two basic processes of peer group dynamics:
peer selection and socialization.
Peer selection processes

Theorizing about the nature of selection processes can be characterized by
one statement: homophily rules. In childhood, as well as in adolescence or
adulthood, friends and self-selected members of peer groups are usually
more similar to one another than they are to people who are not friends or
not members of the group (cf. Cairns et al., 1989; Cohen, 1977; East et al.,
1992; Hallinan & Williams, 1990; Jackson et al., 1991; Kandel, 1978a;
Wright, Giammarino, & Parad, 1986).
Most findings about peer selection processes, when considering both
the friendship and the social network research literatures, suggest that ho-
mophily may go in positive as well as negative directions. Traditionally,
much attention has been paid to how deviant adolescents affiliate with de-
viant others (e.g., Cairns et al., 1988; 1989; Cohen, 1977; Dishion, Patter-
son, Stoolmiller, & Skinner, 1991; Kandel, 1978a). However, there is also
research that focuses on the positive side of peer affiliations, and among
these are studies that include attention to academic characteristics (e.g.,
Cohen, 1977; Hallinan & Williams, 1990; Kandel, 1978a).
How can selection processes be examined? A simple way is to inspect
correlations between scores of individuals and of the members of their
peer group(s). This can be done by using peer profile scores, or by using
intra-class correlations between individuals and the members of their peer
group(s). If group profiles are used, the correlation of individuals' own
scores with the profile of their peer group can be interpreted in analogy to
item-total correlations, in which an individual would be an item, and the
group the total. Thus, these correlations give information about the extent
to which individuals are similar to their peer group members.
Alternative strategies are also possible. In cases in which there is not
much overlap across groups and not many individuals who hold member-
ships in many groups at the same time, analyses of variance can be used in
order to test whether variances within groups are smaller than those across
groups (see Kindermann, 1993). Alternatively, peer group profiles can be
based directly on the variance of the scores of a student's network mem-
bers (instead of their average), and this variance can be compared with the
variance across the other students who were not in this student's peer
group (i.e., forming a nonpeer group profile). Group homogeneity would
again be denoted by differences between the group profiles and the non-
group variances.
Peer socialization processes

Many motivational researchers assume that peers exert socialization influ-
ences on students' motivational development in school (e.g., Dweck &
Goetz, 1978). Peer networks have been labeled "socialization templates"
that define students' opportunities for interactions, opportunities for ob-
serving others in interactions, and their access to situations and activities
(Feiring & Lewis, 1989, p. 125).
Strong socialization effects have been found in studies in which selec-
tion effects were experimentally controlled, as, for example, in the classic
study by Sherif, Harvey, White, Hood, and Sherif (1961), in which chil-
dren were assigned to different groups in a summer camp. However, when
groups of peers were self-selected, rather than experimentally assigned,
indications for socialization effects were found in some studies (e.g., Kan-
del, 1978b), whereas other studies caution us that these effects may be
small (e.g., Berndt et al., 1990), or even negligible compared to selection
processes (e.g., Cohen, 1977).
In the peer relationships literature, notions of socialization processes
usually imply expectations that members of a group become more similar
to one another across time. For example, Hall and Cairns (1984) found ef-
fects of social modeling by peers in an analogy to Bandura's classic bobo-
doll study on aggression. Modeling of a peer was found to have stronger
effects than the experimental manipulations themselves; it was the behav-
ior of the peer that predicted most strongly whether aggression occurred.
With regard to motivation, Berndt and colleagues (1990) found that dis-
cussions among pairs of friends influenced their decisions in motivation-
related dilemmas (e.g., whether to go to a rock concert or to complete a
homework assignment) and made friends more similar in their decisions.
Often, however, socialization researchers are more interested in a fur-
ther hypothesis, namely, that individuals in different groups do not just
change toward the mean of their own group, but change in a way that
magnifies the existing differences between groups. Thus, in a specific
variable under study, the initially "rich" should become "richer" across
time, and the "poor" should become "poorer." Accordingly, in the study by
Berndt and colleagues, initially highly motivated students were expected
to become more motivated, whereas low motivated students were expected
to decrease in motivation. While this hypothesis was not supported in this
specific study, it is nevertheless central to many investigations, especially
to studies focusing on the negative role of peer group affiliations.
How can these socialization expectations be examined? If a students'
peers influence his or her motivational development such that affiliations
with highly motivated peers have a positive effect, while being with disaf-
fected peers has a negative effect, change in this student's motivation
across time should be related to the motivational profile of his or her peer
group at an earlier point in time. Regression analyses can be used to ex-
amine whether students' peer group scores can predict their own engage-
ment at a later point in time, over and above their own earlier engagement.
Significant correlations would indicate that the motivational composition
of a student's peer group is related to change in his or her motivation
across time.
Availability of same-aged peers and selection and

socialization processes in adolescence
Peer selection processes are known to be based primarily on similarity
among candidates. The extent to which candidates are available in a class-
room who are highly similar to target students is an important factor in
these processes. In fact, Kandel (1978a) found that among adolescents,
similarity in grade levels was the one criterion in which friendship dyads
showed most similarity. Although this is not a critical issue in traditional
elementary classrooms, in which children are usually of the same grade or
age and peer group members can be selected from equals, this can be of
major importance in classrooms that include students of different grades
or ages, as is often the case for adolescents.
Thus, if a classroom of adolescents is mainly attended by students
from a lower grade, but only by some students from higher grades, this
can limit the latter students' access to peers who are most similar to
them. Even if many classmates were of the same developmental level as
a target student, or similar to him or her in terms of academic interests,
differences in grade levels or age may nevertheless make these peers
quite inaccessible (or unacceptable) as potential members of that stu-
dent's peer group.
Not much is known about the influences of classroom's age composi-
tion on student development in general, and even less about influences on
academic variables. Most of the available information pertains to child-
hood. While reviews suggest that mixed-age playgroups and classrooms
can have positive effects for young children (Bailey, Burchinal, &
Me William, 1993; Howes & Farver, 1987; Urberg & Kaplan, 1986) and in
later childhood (Miller, 1990; Pratt, 1986), this is much less clear with re-
gard to adolescence. Often, negative expectations prevail with regard to
adolescents who affiliate with older peers (cf. Magnusson, 1988).
Because mixed-age classrooms can impose constraints on peer group
formation processes in terms of availability of most similar others (i.e., of
classmates of the same age), analyses of selection and socialization
processes should be able to incorporate differential expectations in ado-
lescents' classrooms. For example, subgroups of students with different
amounts of mixed-age peer group networks can be compared if sample
sizes are large enough. Alternatively, proportions can be used of students'
peer group members who are of the same age (versus different ages).
These proportions can be used as controls in regression analyses, if it is of
interest whether peer homogeneity exists over and above an average level
of mixed-age affiliations overall. Alternatively, they can be used as

weights if it is of interest whether the amount of same-age (or cross-age)
affiliations contributes to overall peer group homogeneity.
Similarly, the peer groups' age composition can be also taken into con-
sideration for socialization analyses. Usually, controlling for same-age or
cross-age affiliations will be helpful if differential effects are not of inter-
est; weighting procedures will be helpful if expectations exist that cross-
age (or same-age) affiliations exert especially powerful socialization in-
fluences.
Illustration: Peer groups and engagement across the

school year in children and adolescents
The second part of this chapter will illustrate the use of the peer network
identification method and the procedures we use to examine motivational-
ly relevant peer selection and socialization processes. Data from two stud-
ies will be used: a study on 4th and 5th grade children (Kindermann,
1993), and an unpublished study of 9th through 12th grade adolescents.
In terms of its participants, the adolescents' study is quite different
from the children's study. We had two goals with this study: One goal was
to examine the use of the network identification and description methods
in this age range, and to find evidence for similar processes of motivation-
ally based peer selection and socialization processes. The second goal was
to do this in a less homogeneous setting, in which students were from a
more diverse sociodemographic and ethnic background.
In the children study, two 4th and two 5th grade classrooms (n = 115)
were targeted in a rural suburban school district in upstate New York. Stu-
dents were lower-middle to middle class and almost equally divided by
grade, classroom, and gender. The adolescents' study took place in an ur-
ban school district in Oregon, targeting five mixed-grade science classes
(n = 102) that were attended by students from 9th through 12th grade.
Slightly more than half of the students were male, and about half of the
students were 9th graders. The classrooms exhibited a wide range of eth-
nic diversity; about 30% of the students were African American and about
30% of Hispanic or Asian origin.
All adolescents' classes were taught by the same teacher, who had the
explicit goal of encouraging group work and cooperative learning. Many
students in these classes had previously experienced problems with sci-
ence classes; in fact, these classes were part of an effort to restructure the
school's science curriculum. We hoped that in such a setting peer group
structures would be more pronounced, that their relations to classroom en-

gagement would become more clear, and that, by working with one single
teacher, interindividual differences in teachers' standards would be held
constant.
Peer groups at the beginning of the school year

In the 4th and 5th grade classrooms, 57 children were individually inter-
viewed about peer networks at the beginning of the year. In the adoles-
cents' study, 68 students from the five classrooms filled out network as-
sessments in a paper-and-pencil format. In these classrooms, candidates
could also be freely chosen, but in contrast to the children's study, adoles-
cents were encouraged to think about science work-groups in their re-
ports. (Work groups were usually self-selected, with little guidance from
the teacher.) Accordingly, there was some overlap in nominations across
classrooms; however, adolescents did not include peers who did not attend
any of the five science classes. (Figure 12.1 depicts one of the adoles-
cents' classrooms.)
In general, networks of adolescents were more complex than those of
children. For children, although their networks were quite complex in
some classrooms, there were mostly dyads, and larger networks were rare
(nevertheless, there was one with 7 members). For adolescents, although
peer group structures were often larger, they were also more distinct and
less overlapping; there were many dyads and triads, and there was one
cluster that consisted of 14 students. Among both children and adoles-
cents, there were students who were not connected to a peer group (13
children, 11 adolescents).
On average, a child had about 2.2 other students included in his or her
group; group sizes did not differ across grades. In comparison, an adoles-
cent had 3.2 students in his or her group, and group size decreased with
increasing grade levels (from 4.3 for 9th graders, to 1.7 for the 11th and
12th graders combined). Nevertheless, there were no indications of grade
differences in the number of students who were not found with a peer
group.
We had also expected to see the "gender gap" in students' peer groups
decrease with age. For children, groups were exclusively comprised of
peers of the same sex; among adolescents, there were cases in which
groups bridged across genders. Often, just one student of a different sex
was included in an otherwise same-sex group. On average, 80% of an ado-
lescent's peer group members were of the same gender.
Reliability of peer group assessments

Examinations of the extent to which maps of identified groups were con-
sistent with individual students' nominations focused on errors of com-
mission only. Thus, students who knew just about some groups but not
about others were still considered accurate if they agreed with their class-
mates on those accounts that they did report. The overall kappa indices
were .70 for children and .84 for adolescents.
Engagement at the beginning of the school year

Students' school motivation was measured using self-reports as well as re-
ports by their teachers. Connell and Wellborn (1991) have developed a
ten-item scale that assesses students' perceptions of their own behavior in
the classroom (e.g., "I try as hard as I can in school," or "When I'm in
class, I just act like I'm working"). A parallel scale exists for measuring
teacher-perceptions of students' engagement (Skinner et al., 1990; Well-
born, 1991; e.g., "In my class, this student just tries to look busy," or " . . .
works as hard as he/she can"). In previous studies, the scales were shown
to have high internal consistency and stability across a school year (Skin-
ner & Belmont, 1993), to be moderately intercorrelated, and correlated
with grades and achievement scores in mathematics, language, and sci-
ence (Skinner et al., 1990; Wellborn, 1991).
In both studies, students' self-reported engagement was initially mea-
sured about one month after the beginning of the school year; 109 chil-
dren and 90 adolescents participated. With regard to teacher reports of
motivation, the four teachers of the children's classrooms provided reports
for all participating students; in the adolescents' study, the science teacher
reported on 47 students from three of the five classrooms.
We had only partial success in securing a wider range of motivational
scores in the adolescents' data than were obtained with children. On aver-
age, children had a self-reported motivation score of 3.2 (the scale ranges
from 1 to 4), and adolescent's average motivation score was 2.9. Also, the
ranges of individuals' scores were more comparable than we had expected
(for children, they ranged from 1.33 to 4.0; for adolescents from 1.46 to
4.0; ££>:? = .49 and .44).
As was the case in the children's data, the adolescents' teacher also
tended to regard her students' motivation to be significantly lower than
they did themselves. However, adolescents' classrooms showed a smaller
range of engagement scores (SD = .44; average 2.62) than did the chil-
dren's classrooms (SD = .59; average 3.00). This may have been due to the
decision to work with one teacher only. It should be noted that the teacher
had not paid selective attention to specifically high or low motivated ado-
lescents; the 47 students with teacher-reports did not differ in their self-re-
ports from the other students on whom no reports were available.
Grade differences
In the literature, there is ample evidence for a decline in children's motiva-
tion for school with increasing grade level (e.g., Skinner & Belmont,
1993). In the children's study, there had also been differences favoring
younger children (self-report averages: 3.24 for 4th, and 3.07 for 5th
graders). We found further decreases for adolescents (from 3.0 in 9th
grade down to 2.7 for the 1 lth and 12th graders combined). However, as
was true for the children's study, there were no grade differences in ado-
lescents' teacher-reported engagement. Thus, adolescents in higher grades
felt less motivated, but did not appear so to the teacher.
Peer groups and motivation at the beginning

of the school year
Our main questions were about the motivational composition of children's
peer groups in elementary school and of adolescents' peer groups in
mixed-age classrooms. We were first interested in the extent to which stu-
dents were affiliated with others who shared a similar motivational orien-
tation to school. Group similarity is an important indicator that peer selec-
tion processes proceed according to motivational criteria.
Peer group selection

As descriptors for the motivational profiles of students' peer groups, we
used the average of their members. Analyses of variance showed that, for
either children or adolescents, there were no significant differences be-
tween individual students' scores and their peer group scores. As indica-
tors of the extent to which selection processes had led students to select
members for their peer groups who were similar to themselves, we used
correlations between individuals' scores and the scores of their peer net-
work profiles. For both children and adolescents, individuals' own scores
corresponded significantly to the profile scores of their networks with re-
gard to self-reported motivation (with .28 and .27). With regard to teacher
perceptions, group homogeneity was present for children (r = .55), but not
adolescents (r = .23).
Grade differences
There were little grade differences in children's or adolescents' peer
groups. The only exception was found with regard to teacher reports of
adolescents' motivation; 11th and 12th graders' peer groups were per-
ceived as significantly more motivated than groups of students in grades 9
and 10 combined (profile averages 3.25 versus 2.52). This was contrary to
our expectations, because there had been declines with grade in adoles-
cents own self-reported engagement. We concluded that students in higher
grades, although they were less motivated themselves, were nevertheless
affiliated with peers who were not less motivated than the peer group
members of students from lower grades.
With regard to peer group homogeneity, there were no grade differ-
ences for children. Across adolescents' grades however, peer group homo-
geneity was pronounced for the 9th graders (self-report: r = .50, n = 36,
p < .01; teacher-report: r = .78, n = 17, p < .001.), but did not exist in
higher grades. Group similarity seemed to decrease with increasing grade
level in the mixed-grade classrooms. Although students in lower grades
had succeeded in seeking out others who were motivationally similar to
themselves, adolescents in higher grades may not have had the same op-
portunity to do so. This was also evident in the proportions of mixed-
grade versus grade-homogenous peer affiliations. On average, 9th graders
had 90% of their affiliations with other 9th graders, 10th graders had 60%
from the same grade, but 11th and 12th graders had 60% of their peer
group members from grades outside their own.
In sum, there was evidence that children were affiliated with peer
groups that were similar to themselves at the beginning of the school year.
However, for adolescents in the mixed-grade classrooms there were differ-
ences with regard to grade levels. Students in 9th grade were more highly
motivated themselves, usually had larger networks that more often includ-
ed other 9th graders, and there was considerable homogeneity within
these groups. Adolescents in higher grades, who were less motivated
themselves, had fewer classmates with whom they were affiliated, a high-
er percentage from lower grades, and their groups were less homogeneous
with regard to their motivational composition. Nevertheless, it appears
that some of these students were able to compensate for their own low mo-
tivation by affiliating with others who were at least as motivated as the
group members of students from lower grades.
Peer groups across the school year

At the end of the school year, 27 of the children from one 4th grade class-
room were individually reinterviewed about peer networks in the class-
room. In the adolescents' study, 28 students participated again from three
of the five classrooms that were studied initially (n = 61). Although the
number of reports in the adolescent study appears to be low, the reliability
index of the composite map across the individual reports at the end of the
year (K = .88) indicates that the composite map is nevertheless reliable.
All of the children who had been without a group at the beginning of
the year in the one longitudinal 4th grade classroom had acquired mem-
bership in a group by the end of the school year. For adolescents in the
longitudinal classrooms, 6 of the 8 students who had been without a group
at the beginning of the year were still without a group at the end; only two
students had gained membership in a group across the year. However,
there were 19 additional adolescents who had lost all of their initial affili-
ations. (All of these were not present at the second measurement point of
the study.)
Group stability
Across time, stability of children's peer group membership was low; about
50% of a child's peer group members were exchanged. Adolescents'
groups showed even larger instability and a 75% rate of member turnover.
Whereas children's membership changes were mostly due to additions of
new members (on average, group size increased from 2.2 to 2.64 mem-
bers), adolescents' peer group changes were characterized by loss of
members (on average, group size decreased from 3.16 to 1.85 members).
However, this does not imply that change was random; there were some
groups that stayed entirely stable across the year (e.g., the cluster of stu-
dents ALI to EVE in Figure 12.1). Also, stability of adolescents' peer
group memberships decreased significantly with increasing grade level;
for students who were above grade 9, more than 80% of their peer group
members were exchanged or lost across time.
Engagement across the school year

Within a month of the end of the school year, all of the children of the lon-
gitudinal 4th grade classroom participated again in a second questionnaire
assessment. In the adolescent study, 28 students from the three longitudi-
nal classrooms participated again; 23 students were not present during any
of the three days when the survey took place. According to their records,
these students had not dropped out of school but were just absent. Seven
other students were excluded who had left the school or had shifted to dif-
ferent classrooms that were not part of the study; the status of three miss-
ing students could not be determined.
Across the school year, there were no significant changes in children's
or adolescents' motivation. There were also no grade differences in en-
gagement scores at the end of the school year for either children or adoles-
cents. Analyses of variance showed that this was an outcome of differen-
tial attrition in the sample. The students from the longitudinal classrooms
who were present at the second measurement point had already been more
engaged at the beginning of the study in terms of self-reports as well as
teacher-reports.
Peer group selection and socialization processes

across the year
For purposes of examining peer selection processes across the year, we
treated individual students' motivation as if it had remained constant
across the school year. In order to construct peer group profiles for the
end of the school year, we used students' own engagement scores from the
beginning of the school year in combination with the information on their
group affiliations at the end of the year. Thus, correlations between stu-
dents' peer group profiles at the beginning and end of the year give infor-
mation about the extent to which group member turnover had an influence
on the groups' motivational composition across time, when students' in-
traindividual motivational change across time was controlled.
Children's as well as adolescents' peer groups had remained quite sta-
ble across the year in terms of their motivational composition, despite the
fact that at least half of the members had been exchanged. Correlations
between group scores across time were significant for self-reported en-
gagement in children (r = .47; p < .05, n = 25) and adolescents (r = .45,
p < .01, n = 37), as well as for teacher-reports (children: r = .80,/? < .001,
n = 25; adolescents: r = .69, p < .001, n = 31). Thus, although many of the
members of a student's peer network were exchanged, there was neverthe-
less considerable continuity in the motivational make-up of students'
groups.
With regard to adolescents, an additional indication for peer group
continuity across time can be seen in the fact that stable peer groups also
led to a higher probability that students participated in the second mea-
surement point of the study. On average, nearly 40% of the stable mem-
bers of a student's group were present in the classroom again at the end of
the year. However, among students' ties that were not maintained across
the year, 95% of these were associations with peers who were not present
at the end of the year.
Finally, we examined potential socialization influences of students'
peer groups on their own motivational development. For these analyses,
we used students' group profiles from the beginning of the year in combi-
nation with their individual self-reports of engagement from both the be-
ginning and the end of the school year.
For children, there were clear indications of motivational socialization
through peer groups. Regression analyses examined the extent to which
individuals' engagement at the end of the year could be predicted from
their peer group scores at the beginning of the year, when their own en-
gagement at this earlier time was controlled. The results showed that
changes in children's own engagement could be predicted by the initial
composition of their peer networks (p$ = .15, t = 2.06, p < .05, n = 96). For
the adolescent study, we had originally hoped for stronger effects, because
we expected a broader range of self-reported engagement scores. We did
not find such a broad range in our sample. In addition, we needed to re-
vise our expectations, because of the very high rate of membership
turnover that existed overall. Probably, little can be expected in terms of
peer group socialization if the overwhelming majority of one's group
members do not remain stable socialization agents across time.
Hence, we included two further refinements in the analyses. First, so-
cialization effects were expected to be stronger in adolescent peer groups
that stayed together for a longer time, and we included an index of net-
work stability across time for our analyses. This index was the percentage
of individuals who remained stable members of a student's group across
time, and was used as a control. Thus, socialization effects were examined
under the assumption that all students had the same number of stable peer
group members. By the same token, adolescents who did not have at least
one stable member were excluded from the analysis.
Secondly, we took the mixed-grade design of the adolescent class-
rooms into consideration. Students' grade differences were of less interest
to us than the question of whether the mixed-grade design of the class-
rooms offered opportunities for some students in terms of their motiva-
tional development. Since some (but not all) of the peer groups in our
adolescent classrooms consisted of students from different grades, we
wanted to examine whether the grade-composition of students' peer net-
works contributed to changes in students' engagement in combination
with the motivational characteristics of these groups. In order to do so, we
included the proportion of an individuals' peer group members that were

in different grades as a weighting factor in the analysis.
Thus, we examined whether an individual student's peer group profile
at the beginning of the year, when weighted with the proportion of his or
her peer group members who were in a different grade, and controlling for
the level of stability within his or her group, allowed us to predict how that
student's own motivation would change across time. The results were con-
sistent with our expectations: Students who were with highly motivated
peer groups and who had many of their members from grades that were
different from their own were likely to increase in motivation across time
(on average by about .20 points on the 4-point scale). Students who were
affiliated with groups that were low on motivation and showed little grade
diversity decreased slightly across time ((3 = .48; t = 2.51, n = 18,/? < .05).
We regard this as an indication that the mixed-grade design of the class-
rooms (as well as the teacher's encouragement to form groups that work
together) had indeed offered some advantages for those students who
managed to bridge across grade barriers in their peer networks.
Discussion: Peer group selection and socialization processes

in elementary school and high school classrooms
The goal of this chapter was to describe a method for identifying students'
peer groups in school as well as its applications for studying peer selec-
tion and socialization processes in childhood and adolescence. Specifical-
ly, we wanted to examine the use of this method in a setting of adolescents
that was characterized by a large amount of ethnical diversity. In adoles-
cence, classrooms are quite common in which students differ with regard
to age and grade level, although these classrooms may be more homoge-
neous with regard to students' performance or academic achievement. As
was the case in the current study, these classroom environments are often
formed with the specific goal of encouraging students' to learn from one
another, both in terms of social and academic development.
The results of both studies indicated that at the beginning of the school
year, self-selection processes among the students led to peer groups that
were quite homogeneous in terms of their motivational composition. The
major difference between children and adolescents was that there were
grade differentiations in adolescents' (mixed-grade) classrooms. Peer
group homogeneity was stronger for 9th graders and for students who had
more peer group members from their own grade.
Across the school year, children's peer groups remained moderately
stable in terms of peer group memberships, whereas adolescents' groups

were characterized by an enormous turnover rate. This rate of change
matched perceptions of the teacher, who also reported that adolescents'
groups changed greatly in terms of work group membership. Neverthe-
less, the motivational make-up of students' peer groups remained quite
stable across time. Peer group members seem to have been replaced,
dropped, or newly integrated into networks in a way that left the motiva-
tional composition of these groups intact.
Peer socialization processes were examined as the extent to which the
motivational profile of a student's peer group members at the beginning of
the school year allowed for the prediction of that child's own motivational
change across time. For children, we found clear evidence that self-select-
ed peer group contexts can have socializing effects on individuals across
time: Children who were affiliated with highly motivated groups changed
positively across the school year, while children who were with less moti-
vated groups changed negatively. For adolescents, the evidence for social-
ization effects was clearer for those students who had peer networks that
included many peers from different grades. Although this needs to be tak-
en with caution, because our survivor sample of students who participated
at both measurement times had been relatively highly motivated right
from the start, the results are nevertheless encouraging. Despite the differ-
ences, there are clear similarities with regard to the general finding that
the motivational composition of students' peer groups at the beginning of
the school year was indicative of these student's own motivational devel-
opment across the year.
School motivational and peer selection and socialization processes.

Because these findings are correlational, there can be doubts whether stu-
dents' motivational characteristics can really be regarded to be direct tar-
gets of selection and socialization processes. What appear to be indica-
tions of selection and socialization processes according to motivational
criteria may be by-products of processes that do not directly target school
motivation, but are directed toward other criteria that are more salient in
interactions among students.
Indeed, studies by Cohen (1977) and Kandel (1978a) caution us that
academic variables may not be the characteristics on which peer group
members may be most similar to one another. For example, selection
processes may be more based on sociodemographic variables, achieve-
ment, intelligence, or on classroom behavior that is openly observable by
others, at least at the beginning of a school year. Similarly, socialization
processes may also not target motivation directly, but rather academic or
social behaviors. Some of these behaviors may be facilitative of (or com-

patible with) classroom engagement and others may be incompatible with
classroom engagement. Considering that there are relations between stu-
dents' social behavior in the classroom and their academic achievement
(cf. DeBaryshe et a l , 1993; Wentzel, 1991; Wentzel, 1993), it may well be
that peer selection and socialization processes are more directed towards
social behaviors, rather than academic motivation.
The extent to which this is the case may be a matter of the specific
classroom settings in which peer processes are studied, of the classrooms'
agenda, and of the specific goals of the teacher, but also of the kinds of
students who attend the classrooms. For example, it seems likely that stu-
dents' age can play a role in the extent to which peer groups are influential
for their own motivational development. However, in our understanding,
all of these variables will mostly affect the strength with which peer selec-
tion or socialization processes pertain to school motivation. In other
words, if motivational change is a by-product of peer group processes,
links to peer selection and socialization processes can be expected to be
weaker than if motivation was a direct target of these processes.
Integrating the strands of peer relationships research

Our studies point to peer group selection processes as key processes by
which children and adolescents seem to select contexts for themselves in
the classroom, which then, via socialization processes, have implications
for their own further motivational development. On this general level, our
results seem to be entirely compatible with results of other researchers,
such as, for example, studies of children's or adolescents' friendship pat-
terns. Nevertheless, we think that the current framework can provide some
additional information.
The methods presented for peer group identification seem to be espe-
cially promising if they are combined with existing frameworks from the
friendship and sociometrics literatures. By combining network methods
with sociometric classification systems (e.g., Coie et al., 1982; Newcomb
& Bukowski, 1983), we could find out about the role of sociometric popu-
larity within and across peer groups. Do individuals who belong to popu-
lar, neglected, or rejected categories of children share associations with
each other, or do they also belong to different parts of the social world in
the classroom? Are popular children "stars" who have connections to all
kinds of different other children, whereas rejected children are more iso-
lated? Are neglected children "satellites," and unlikely candidates for
group inclusion who have outlier positions in otherwise more coherent
networks? Are the children within one's network perceived to be more

likeable than children on the outside?
Further questions could also be addressed by combining network
methods with methods used in the friendship literature (e.g., Ladd, 1990).
Are one's friends usually members of one's network, or is friendship
something special, and one's close friends are as likely to be members of
one's peer network as they are to be outsiders? Are friends more influen-
tial in terms of socializing influences than the larger group of one's peer
network members?
Combining these methods could also help in addressing more specific
questions about the motivational relevance of the processes under study.
Processes of how students select members for work groups in school may
be affected by the peers' overall popularity, and so may the peers' socializ-
ing influences on individuals. Work groups that include one's friends may
have stronger socializing influences than work groups that do not. In par-
ticular, a student's peers may have influences on the formation or mainte-
nance friendships, and it is entirely possible that resulting friendships have
socializing influences that are stronger and more specific than those that
emanate from network members who do not become this student's close
friends.
Peer group networks as socialization contexts for

motivational development
We have started our chapter by pointing out that contemporary research
on school motivation is mainly oriented toward intraindividual explana-
tions, or toward explanations that focus on the teacher as the central moti-
vating force. In contrast, our framework focuses on students' social rela-
tionships with peers, and on processes of peer selection and socialization.
Because peers are students themselves who are developing at the same
rate as target students under study, it is essential that a framework for
studying their influences pays attention to the notion that individuals' de-
velopment proceeds within contexts that change themselves, and that indi-
viduals can have an active role in determining who or what will be a so-
cialization context for their development (Kindermann & Skinner, 1992;
Kindermann & Valsiner, 1995).
On a large scale, the methods presented in this chapter have two gen-
eral objectives. The first goal was to show that it is possible to par-
tition the social ecology of a student in the classroom into sub-units that
can be expected to be particularly influential contexts for this student's
further development. The second goal was to demonstrate how specific

pathways of reciprocal person-context influences can be examined across
time.
Findings that children's and adolescents' peer groups change rapidly
across the time span of a school year can make it difficult to study their
influences. A large amount of member turnover in students' peer groups
across time can be regarded as a sign that nothing stays stable in these
contexts, and that changes are rather unpredictable. We regard our find-
ings as indicating that this depends on how we conceive students' peer
contexts in school. While the individual "faces" of students' peer net-
works may change rapidly, there can nevertheless be considerable conti-
nuity in the psychological characteristics of these groups. In our findings,
not only did the motivational profiles of children's and adolescents' peer
networks remain quite unaffected by vast changes in who was a member
of these groups at different points in time; in addition, there also were in-
dications that the motivational characteristics of students' were indicative
of students' own motivational change across time.
The methods presented in this chapter seem to offer ways to identify
contextual agents as well as to examine their developmentally influential
characteristics at different points in time. These methods also seem to al-
low us to examine how changes within individuals are related to changes
within their contexts. The specifics of these methods may be in need of
further refinement and elaboration. However, for studies of the role of
self-selected peer networks on student's development, we think that it is
this perspective on developing person-context relationships, over and
above the specifics of the methods used, that may be most useful for fu-
ture research on the social determinants of students' school adjustment
and motivation.
Acknowledgment
We want to thank Robert B. Cairns from the University of North Carolina
at Chapel Hill, who developed the network assessment method on which
this chapter is based, and Ellen A. Skinner from Portland State University
and James P. Connell from the Institute for Research and Reform in Edu-
cation, Philadelphia, PA, for methodological advice. We also want to
thank the students who participated in the studies. Special thanks go to
Dawn Oostman, the teacher of the adolescents' classrooms, Christine
Borgford from Portland State University, Joan E. Crosby, Vice-Principal,
and Fred Rectanus, Coordinator of the Institute for Science and Mathe-
matics at Grant High School for their support for the study. Matthew B.
Hall from Portland State University and Meredith Boatsman from Lewis
& Clark College deserve our thanks for their help with the data collection.
Note
This research was supported by a Faculty Development Grant from Port-
land State University, and a grant from the National Institutes of Health
(NICHD, 1R15HD31687-01).
References
Adams, R. G. (1989). Conceptual and methodological issues in studying friend-
ships of older adults. In R. G. Adams and R. Blieszner (Eds.), Older adult
friendship: Structure and process (pp. 17^41). Newbury Park, CA: Sage.
Ainley, M. D. (1993). Styles of engagement with learning: Multidimensional as-
sessment of their relationship with strategy use and school achievement. Jour-
nal of Educational Psychology, 85, 395-405.
Ames, C. (1984). Competitive, cooperative, and individualistic goal structures: A
cognitive-motivational analysis. In R. E. Ames & C. Ames (Eds.), Research on
motivation in education (vol. 1. pp. 177—207). Student motivation. Orlando,
FL: Academic Press.
Ames, C , & Ames, R. E. (Eds.), (1985). Research on motivation in education.
(vol. 2). The classroom milieu. Orlando, FL: Academic Press.
Ames, R. E., & Ames, C. (Eds.), (1984). Research on motivation in education.
Student motivation (vol. 1). Orlando, FL: Academic Press.
Asher, S. R., & Coie, J. D. (Eds.) (1990). Peer rejection in childhood. New York:
Cambridge University Press.
Asher, S. R., & Hymel, S. (1981). Children's social competence in peer relations:
Sociometric and behavioral assessment. In J. D. Wine & M. D. Smye (Eds.), So-
cial competence (pp. 125—157). New York: Guilford.
Bailey, D. B., Burchinal, M. R., & McWilliam, R. A. (1993). Age of peers and ear-
ly childhood development. Child Development, 64, 848-862.
Bakeman, R., & Gottman, J. M. (1986). Observing interaction: An introduction to
sequential analysis. Cambridge, England: Cambridge University Press.
Bennett, R. E., Gottesman, R. L., Rock, D. A., & Cerullo, F. (1993). Influence of
behavior perceptions and gender on teacher's judgments of students' academic
skill. Journal of Educational Psychology, 85, 347-356.
Berndt, T. J. (1989). Friendships in childhood and adolescence. In W. Damon
(Ed.), Child development today and tomorrow (pp. 332—348). San Francisco,
CA: Jossey-Bass.
Berndt, T. J., Laychak, A. E., & Park, K. (1990). Friends' influence on adoles-
Boggiano, A. K., & Katz, P. (1991). Maladaptive achievement patterns in stu-
dents: The role of teachers' controlling strategies. Journal of Social Issues, 47,
35-51.
Breiger, R. L. (1988). The duality of persons in groups. In B. Wellman & S. D.
Berkowitz (Eds.), Social structures: A network approach (pp. 83-98). New
York: Cambridge University Press.
Brophy, J. (1983). Research on the self-fulfilling prophecy and teacher expecta-
tions. Jo urnal of Educational Psychology, 70, 154-166.
Brophy, J. (1985). Teacher's expectations, motives, and goals for working with
problem students. In C. Ames & R. E. Ames (Eds.), Research on motivation in
education The classroom milieu, (vol. 2, pp. 175-214). Orlando, FL: Academic
Press.
Brophy, J. (1986). Teacher influences on student achievement. American Psychol-
ogists, 41, 1069-1077.
Buhrmester, D. (1990). Intimacy of friendship, interpersonal competence, and ad-
justment during preadolescence and adolescence. Child Development, 61,
1101-1111.
Cairns, R. B., Cairns, B. D., & Neckerman, J. (1989). Early school dropout: Con-
figurations and determinants. Child Development, 60, 1437-1452.
Cairns, R. B., Cairns, B. D., Neckerman, H. J., Gest, S. D., & Gariepy, J.-L.
(1988). Social networks and aggressive behavior: Peer support or peer rejec-
tion? Developmental Psychology, 24, 815-823.
Cairns, R. B., Gariepy, J. L., & Kindermann, T. A. (1990). Identifying social clus-
ters in natural settings. Unpublished manuscript: University of North Carolina
at Chapel Hill.
Cairns, R. B., Perrin, J. E., & Cairns, B. D. (1985). Social structure and social cog-
nition in early adolescence: Affiliative patterns. Journal of Early Adolescence,
5, 339-355.
Chapman, M., Skinner, E. A., & Baltes, P. B. (1990). Interpreting correlations be-
tween children's perceived control and cognitive performance: Control, agency,
or means-ends beliefs? Developmental Psychology, 23, 246-253.
Cohen, A. K. (1977). Sources of peer group homogeneity. Sociology of Educa-
tion, 50,221-241.
Coie, J. D., Dodge, K. A., & Coppetelli, H. (1982). Dimensions and types of so-
cial status. Child Development, 59, 815-829.
Connell, J. P. (1990). Context, self, and action: A motivational analysis of self-sys-
tem processes across the life-span. In D. Cicchetti & M. Beeghly (Eds.), The
self in transaction: Infancy to childhood (pp. 61-97). Chicago, IL: University
of Chicago Press.
Connell, J. P., Spencer, M. B., & Aber, J. L. (1994). Educational risk and resilience
in African-American youth: Context, self, action, and outcomes in school.

Sroufe (Eds.), Minnesota symposium on child psychology, (vol. XXIII, pp.
Corno, L., & Rohrkemper, M. M. (1985). The intrinsic motivation to learn in
classrooms. In C. Ames, & R. E. Ames (Eds.), Research on motivation in edu-
cation (vol. 2, pp. 53-90). The classroom milieu. Orlando, FL: Academic Press.
Covington, M. V (1984). The motive for self-worth. In R. E. Ames & C. Ames
(Eds.), Research on motivation in education (vol. 1, pp. 77-113). Student moti-
vation. Orlando, FL: Academic Press.
Czikszentmihalyi, M , & Larson, R. (1984). Being adolescent. New York: Basic
Books.
DeBaryshe, B. D., Patterson G. R., & Capaldi, D. M. (1993). A performance mod-
el for academic achievement in early adolescent boys. Developmental Psychol-
ogy, 29, 794-804.
Deci, E. L., & Ryan, R. M. (1985). Intrinsic motivation and self-determination in
human behavior. New York: Plenum.
Dishion, T. J. (1990). The family ecology of boys' peer relations in middle child-
hood. Child Development, 61, 874-892).
Dishion, T. 1, Patterson, G. R., Stoolmiller, M., & Skinner, M. L. (1991). Family,
school, and behavioral antecedents to early adolescent involvement with antiso-
cial peers. Developmental Psychology, 27, 172-180.
Dweck, C. S., & Elliott, E. S. (1983). Achievement motivation. In P. Mussen (Se-
ries. Ed.) & E. M. Hetherington (Volume Ed.), Handbook of child psychology
(vol. 4, pp. 103-196). Socialization, personality, and social development. New
York: Wiley.
Dweck, C. S., & Goetz, T. (1978). Attributions and learned helplessness. In J. Har-
vey, W. Ivkes, & R. Kidd (Eds.), New directions in attribution research (pp.
East, P. L., Lerner, R. M., Lerner, J. V, Soni, R. T, Ohannessian, C. M., & Jacob-
son, L. P. (1992). Early adolescent-peer group fit, peer relations, and psychoso-
cial competence: A short-term longitudinal study. Journal of Early Adoles-
cence, 12, 132-152.
Eccles, J., Midgley, C , & Adler, T. F. (1984). Grade-related changes in the school
environment: Effects on motivation. In J. G. Nicholls (Ed.), The development of
achievement motivation (pp. 283-331). Greenwich, CT: JAI.
Feiring, C , & Lewis, M. (1989). The social networks of girls and boys from early
through middle childhood. In D. Belle (Ed.), Children s social networks and so-
cial support (pp. 119-150). New York: Wiley.
Furman, W. (1989). The development of children's social networks. In D. Belle

(Ed.), Children s social networks and social support (pp. 151-172). New York:
Wiley.
Gottfried, A. E., Fleming, J. S., & Gottfried, A. W. (1994). Role of parental moti-
vational practices in children's academic intrinsic motivation and achievement.
Journal of Educational Psychology, 86, 104—113.
Graham, S., & Barker, G. P. (1990). The downside of help: An attributional-devel-
opmental analysis of helping behavior as a low-ability cue. Journal of Educa-
tional Psychology, 82, 7—14.
Green, K. D., Forehand, R., Beck, S. J., & Vosk, B. (1980). An assessment of the
relationship among measures of children's social competence and children's
academic achievement. Child Development, 51, 1149-1156.
Grolnick, W. S., & Ryan, R. M. (1987). Autonomy in children's learning: An ex-
perimental and individual difference investigation. Journal of Personality and
Grolnick, W. S., Ryan, R.M., & Deci, E. L. (1991). Inner resources for school
achievement: Motivational mediators of children's perceptions of their parents.
Hall, W. M., & Cairns, R. B. (1984). Aggressive behavior in children: An out-
come of modeling or social reciprocity? Developmental Psychology, 20, 739-
745.
Hallinan, M. T, & Williams, R. A. (1990). Students' characteristics and the peer-
influence process. Sociology of Education, 63, 122-132.
Hartup, W W. (1978). Children and their friends. In H. McGurk (Ed.), Childhood
social development (pp. 181-271). London: Methuen.
Hartup, W. W. (1983). Peer relations. In P. Mussen (Series. Ed.) & E. M. Hether-
ington (Volume Ed.), Handbook of child psychology (vol. 4, pp. 103-196). So-
cialization, personality and development..NewYork: Wiley.
Holloway, S. D., & Hess, R. D. (1982). Causal explanations for school perfor-
mance: Contrasts between mothers and children. Journal of Applied Develop-
mental Psychology, 3, 319—327.
Howes, C , & Farver, J. (1987). Social pretend play in a 2-year old: Effects of age
of partners. Early Childhood Research Quarterly, 2, 305-314.
Hymel, S., & Rubin, K. H. (1985). Children with peer relationships and social
skills problems: Conceptual, methodological, and developmental issues. In G.
Whitehurst (Ed.), Annals of child development (vol. 2, pp. 251-297). Green-
wich, CT: JAI.
Jackson, S., Brett, J. F., Sessa, V I., Cooper, D. M., Julin, J. A., & Peyronnin, K.
(1991). Some differences make a difference: Individual dissimilarity and group
heterogeneity as correlates of recruitment, promotions, and turnover. Journal of
Applied Psychology, 76, 675-689.
Johnson, D. W, & Johnson, R. (1985). Motivation processes in cooperative, com-
petitive, and individualistic learning situations. In C. Ames & R. E. Ames
(Eds.), Research on motivation in education. The classroom milieu, (vol. 2, pp.

249-288). Orlando, FL: Academic Press.
Jussim, L. (1989). Teacher expectations, self-fulfilling prophecies, perceptual bi-
ases, and accuracy. Journal of Personality and Social Psychology, 57, 469-
480.
Kandel, D. B. (1978a). Similarity in real-life adolescent friendship pairs. Journal
of Personality and Social Psychology, 36, 306—312.
Kandel, D. B. (1978b). Homophily, selection, and socialization in adolescent
Keller, J. (1983). Motivational design of instruction. In C. M. Reigeluth (Ed.), In-
structional design theories and models: An overview of their current status.
(pp. 383^34). Hillsdale, NJ: Erlbaum.
Kindermann, T. A. (1993). Natural peer groups as contexts for individual develop-
ment: The case of children's motivation in school. Developmental Psychology,
29, 970-977.
Kindermann, T. A., & Skinner, E. A. (1992). Modeling environmental develop-
ment: Individual and contextual trajectories. In J. B. Asendorpf & J. Valsiner
(Eds.), Stability and change in development: A study of methodological reason-
ing (pp. 155-190). Newbury Park, CA: Sage.
Kindermann, T. A., & Valsiner, J. (1995). (Eds.) Development of person-context
relations. Hillsdale, NJ: Erlbaum.
Ladd, G. W. (1990). Having friends, keeping friends making friends, and being
ment. Child Development, 61, 108-1100.
Larson, R., Richards, M. H. (1991). Daily companionship in late childhood and
early adolescence: Changing developmental contexts. Child Development, 62,
284-300.
Magnusson, D. (1988). Individual development from an interactional perspective.
Mclver, D. J., Stipek, D. J., & Daniels, D. H. (1991). Explaining within-semester
changes in student effort in junior high school and senior high school courses.
Journal of Educational Psychology, 83, 201—211.
Meece, J. L., & Holt, K. (1993). A pattern analysis of students' achievement goals.
Midgley, C, Feldlaufer, H., & Eccles, J. (1989). Change in teacher efficacy and
student self-and task related beliefs in mathematics during the transition to ju-
nior high school. Journal of Educational Psychology, 81, 247-258.
Miller, B. A. (1990). A review of the quantitative research on multigrade instruc-
tion. Research in Rural Education, 7, 1-8.
Moely, B. E., Hart, S. S., Leal, L., Santulli, K. A., Rao, N., Johnson, T, & Hamil-
ton, L. B. (1992). The teacher's role in facilitating memory and study strategy
development in the elementary school classroom. Child Development, 63,
653-672.
Moreno, J. L. (1934). Who shall survive? A new approach to the problem of hu-
man interrelations. Washington, DC: Nervous and Mental Disease Publishing.
Newcomb, A. R, & Bukowski, W. M. (1983). Social impact and social preference
as determinants of peer group status. Developmental Psychology, 19, 856-867.
Newcomb, A. R, Bukowski, W. M., & Pattee, L. (1993). Children's peer relations:
A meta-analytic review of popular, rejected, neglected, controversial, and aver-
age sociometric status. Psychological Bulletin, 113, 99—128.
Nicholls, J. G. (1984). Concepts of ability and achievement motivation. In R. E.
Ames & C. Ames (Eds.), Research on motivation in education. Student motiva-
tion (vol. 1, pp. 39-73). Orlando, PL: Academic Press.
Patrick, B. C, Skinner, E., A., & Connell, J. P. (1993). What motivates children's
behavior and emotion? Joint effects of perceived control and autonomy in the
academic domain. Journal of Personality and Social Psychology, 65, 781-791.
Pintrich, P. R., & De Groot, E. V (1990). Motivational and self-regulated learning
components of classroom performance. Journal of Educational Psychology,
82, 33-50.
Pratt, D. (1986). On the merits of multiage classrooms. Research in rural educa-
tion^, 111-115.
Price, J. M., & Dodge, K. A. (1989). Peers' contributions to children's social mal-
adjustment: description and intervention. In T. J. Berndt & G. WT. Ladd (Eds.),
Peer relationships in child development (pp. 341—370). New York: Wiley.
Sackett, G. P., Holm, R., Crowley, C, & Henkins, A. (1979). A PORTRAN pro-
gram for lag sequential analysis of contingency and cyclicity in behavioral in-
teraction data. Behavior Research Methods & Instrumentation, 11, 366-378.
Schiefele, U. (1991). Interest learning and motivation. Educational Psychologist,
26, 299-324.
Schunk, D. H. (1991). Self-efficacy and academic motivation. Educational Psy-
chologist, 26, 207-232.
Sherif, M., Harvey, O. J., White, B. J., Hood, W. R., & Sherif, C. W. (1961). Inter-
group conflict and cooperation: The robbers cave experiment. Norman, OK:
Institute of Group Relations.
Skinner, E. A., Wellborn, J. G., & Connell, J. P. (1990). What it takes to do well in
school and whether I've got it: The role of perceived control in children's en-
gagement and social achievement. Journal of Educational Psychology, 82,
22-32.
Stevenson, H. W, Lee, S. Chen, C, Lummis, M., Stigler, J., Pan, L., & Ge, R
(1990). Mathematics achievement of children in China and the United States.
Stipek, D. J. (1993). Motivation to learn. Boston, MA: Allyn and Bacon.

chology, 46, 561-574.
Urberg, K. A., & Kaplan, M. G. (1986). Effects of classroom age composition on
the play and social behaviors of preschool children. Journal ofApplied Devel-
opmental Psychology, 7,403—415.
von Eye, A. (1990). Introduction to conflgural frequency analysis: The search for
types and antitypes in cross-classifications. New York: Cambridge University
Press.
Wasserman, S., Galaskiewicz, J. (Eds.), (1994). Advances in social network
analysis. Thousand Oaks, CA: Sage.
Weiner, B. (1979). A theory of motivation for some classroom experiences. Jour-
nal of Educational Psychology, 71, 3-25.
Weiner, B. (1985). An attribution theory of achievement and emotion. Psycholog-
ical Bulletin, 92, 548-573.
Weiner, B. (1986). An attributional theory of motivation and emotion. New York:
Springer.
Weiner, B. (1990). History of motivational research in education. Journal of Edu-
cational Psychology, 82, 616—622.
Weisz, J. R., & Cameron, A. M. (1985). Individual differences in the student's
sense of control. In C. Ames & R. E. Ames (Eds.), Research on motivation in
education. The classroom milieu, (vol. 2, pp. 93-140). Orlando, FL: Academic
Press.
Wellborn, J. G. (1991). Engaged vs. disaffected action: Conceptualization and
measurement of motivation in the academic domain. Unpublished Doctoral
Dissertation, Graduate School of Human Development and Education.
Rochester, NY: University of Rochester.
Wellman, B., & Berkowitz, S. D. (Eds.), (1988). Social structures: A network ap-
proach. Cambridge, England: Cambridge University Press.
tional Psychology, 81, 131-142.
Wentzel, K. R. (1991). Relations between social competence and academic
(2), 357-364.
Wigfield, A., & Karpathian, M. (1991). Who am I and what can I do? Children's
self-concepts and motivation in academic situations. Educational Psychologist,
26, 233-262.
Wright, J. C, Giammarino, M., & Parad, H. W. (1986). Social status in small
groups: Individual-group similarity and the social "misfit". Journal of Person-
ality and Social Psychology, 50, 523-536.
13 Academic failure and school dropout:
The influence of peers
Shelley Hymel, Colin Comfort, Kimberly Schonert-Reichl,

and Patricia McDougall
Academic failure and school dropout pose serious obstacles to the pursuit
of educational success and represent a loss for both the individual and so-
ciety. Recent data indicate that in the United States in 1992, about 11% of
individuals age 16-24 had not completed high school, representing ap-
proximately 3.4 million individuals (Center for Education Statistics,
1993). In urban areas such as Chicago, the dropout rate can reach as high
as 50% for ethnic minority students (Hahn, 1987). In Canada, an estimat-
ed 30% of 15- to 20-year-olds do not complete high school, as compared
with an estimated dropout rate of less than 10% in Germany, and less than
2% in Japan (Employment and Immigration, 1990; Statistics Canada,
1993). The consequences of early school leaving are quite negative, as
dropouts are more likely to experience unemployment and acquire less se-
cure and satisfying work than graduates (McCaul, Donaldson, Coadarci,
& Davis, 1992; Rumberger, 1987). Biemiller and Meichenbaum (1993)
and Catterall (1985) remind us that the existing dropout rate also has di-
rect implications for society in general, not only in terms of loss of poten-
tial of these individuals as contributors to our society, but also in terms of
the cost incurred from unemployment, welfare and assistance programs,
housing, health care, and so on.
Although studies of the causes of school dropout have identified a
wide range of contributing factors, institutional as well as individual, the
primary emphasis in this literature has been on academic and familial fac-
tors. Far less attention has been given to the role of social factors in con-
tributing to and/or protecting against school failure and dropout. Accord-
ingly, in this chapter, we consider the impact of social factors on the
likelihood of academic failure and school dropout, with particular interest
in the role of peers. Our focus on the peer group is based on an extension
of a recent model of school motivation proposed by Ryan and Powelson
(1991), who emphasize that feelings of relatedness (among other things)
313
are critical to education and learning. Although relatedness between stu-

dents and their parents and teachers has been emphasized in research to
date, we suggest that students' relations with peers provide another criti-
cal, but often overlooked, sphere of influence. We begin with a brief
overview of the extant literature on school dropout. Against this backdrop,
we then review research on the role of peers in early school withdrawal,
considering four distinct aspects of peer influence: (1) prior social accep-
tance and rejection, (2) social isolation versus involvement, (3) the nega-
tive influence of peers, and (4) aggression and antisocial behavior. View-
ing dropout as a final act in a gradual process of school disengagement,
we end with a discussion of how peer influences may interact with other
variables to contribute ultimately to the likelihood of school withdrawal
versus completion.
Correlates and predictors of school dropout

To date, research on the correlates of school dropout has focused on
characteristics of the student, particularly academic and familial factors,
with the implication that the causes of dropout must lie primarily in the
scholastic failure of the individual or difficulties within the family. For
instance, students who drop out are more likely to exhibit lower levels of
intellectual ability (Combs & Cooley, 1968; Howell & Frese, 1982;
Lloyd, 1978), poorer achievement and grades (Barrington & Hendricks,
1989; Dryfoos, 1990; Ekstrom, Goertz, Pollack, & Rock, 1986; Ens-
minger & Slusarcick, 1992; Statistics Canada, 1993), higher rates of tru-
ancy (Wehlage & Rutter, 1986), grade retention (Cairns, Cairns, &
Neckerman, 1989; Center for Education Statistics, 1993; Ekstrom et al.,
1986; Howell & Frese, 1982; Statistics Canada, 1993) and school trans-
fers (Stroup & Robins, 1972). With regard to familial variables, dropout
rates are found to be disproportionately higher among lower income
families and some racial minorities (Hahn, 1987), although research by
Rumberger (1983) indicates that the effect of race/ethnicity decreases
when socioeconomic level is held constant, and recent U.S. data indicate
that dropout rates do not differ significantly across ethnic groups (Cen-
ter for Education Statistics, 1993). Dropouts are also more likely to
come from single-parent and no-parent homes (Rumberger, 1983; Statis-
tics Canada, 1993), from homes that provide less support for education-
al success (Ekstrom et al., 1986; Howell & Frese, 1982), less involve-
ment in parent-school organizations, and less strict rules regarding
Academic failure and school dropout 315
school (Ensminger & Slusarcick, 1992). As well, parents of dropouts

tend to exhibit lower levels of educational attainment (Lloyd, 1978; Sta-
tistics Canada, 1993).
Academic and familial factors certainly play a significant role in
school dropout. However, we are concerned that this primary focus within
the literature reflects (perhaps implicitly) an individual deficit model of
school failure that emphasizes characteristics of the student (e.g., socioe-
conomic status, family characteristics, learning problems, absenteeism) as
the major "cause" of academic failure and early school leaving. Moreover,
these factors do not represent a complete picture of the problem, and
alone cannot account for the current rate of school dropout. Even data
used to support the influence of familial and academic factors suggests
that these variables only account for a portion of the dropouts surveyed.
For example, although school leavers are found to perform more poorly in
school than graduates (as reflected in reported grades), most dropouts per-
formed satisfactorily in school and only some reported problems with
school work as a primary reason for leaving school (Center for Education
Statistics, 1993; Statistics Canada, 1993). With regard to familial vari-
ables, a recent study comparing school leavers and graduates indicated
that most dropouts (61%) as well as graduates (83%) came from two-par-
ent homes (Statistics Canada, 1993). Moreover, although dropout rates are
higher among low income families, the majority of dropouts come from
middle income families (Center for Education Statistics, 1993). Thus, the
individual factors which have been highlighted in previous research do not
appear to characterize the majority of dropouts surveyed. As well, many
of the individual variables identified as predictors of school leaving are
nonmanipulable (e.g., SES) and thus do not easily lend themselves to in-
tervention efforts. It therefore becomes imperative to consider a broader
range of factors that also play a role in the decision to drop out of school
early.
What reasons do students give for dropping out? This question be-
comes crucial in light of concerns expressed by Zaslow and Takanishi
(1993) that much of the extant research on adolescence has bypassed the
step of giving students a "voice" and has failed to consider students' own
perceptions of their school experiences. Recent studies in both the U.S.
(Center for Education Statistics, 1993) and Canada (Statistics Canada,
1993) indicate that students' reasons for dropping out of school are quite
varied. In addition to academic and familial concerns, dropouts also high-
light job- or work-related concerns, the desire to travel, and drug/alcohol
problems as reasons for dropout. It is important to note, however, that dis-

satisfaction with the school milieu and other school-related problems top
the list of reasons for leaving school early (Goertz, Ekstrom, & Rock,
1991; O'Sullivan, 1990). Indeed, about 40% of dropouts cite school-relat-
ed concerns and not liking school as a major factor in dropping out (Cen-
ter for Education Statistics, 1993; Statistics Canada, 1993).
Because such a large proportion of dropouts attribute leaving school
early to school concerns, the critical question becomes "what about school
was disliked?" To date, the most obvious answer has been that students
dislike the work they are required to do in school. Still, fewer than one-
third of recent U.S. dropouts indicated not being able to keep up with
schoolwork as a reason for dropping out (Center for Education Statistics,
1993). Dislike of school and school work can also manifest itself in gener-
alized reports of boredom (Farrell, 1990), as students question the person-
al relevance of school for their daily lives. A second, but often overlooked,
reason for disliking school involves the social climate of the school. A
substantial proportion of recent dropouts suggested that they could not get
along with teachers (22.8%) or other students (14.5%), that they did not
feel that they belonged (24.2%), or that they did not feel safe at school
(6.0%) (Center for Education Statistics, 1993). Another 8% indicated that
their friends had dropped out. These data suggest that a substantial num-
ber of dropouts do not feel a sense of connectedness or belonging within
the school context. Thus, it is important to consider how aspects of the ed-
ucational system, in addition to academic and familial difficulties, con-
tribute to the likelihood of school dropout.
In an effort to broaden our understanding of the factors that underlie
school dropout, some researchers have examined aspects of the school
system which may, in part, contribute to early withdrawal. For example,
Fine (1991) suggests that many so-called "dropouts" are better labeled
"pushouts," since they have been "encouraged" to leave school early. As
well, Crespo and Michelena (1981) suggest that the school practice of
streaming students into varying tracks or levels is significantly predictive
of dropout, even after controlling for intellectual ability, academic perfor-
mance, and age. There is also evidence to suggest that dropout rates are
lower in smaller as opposed to larger schools (Pittman & Haughwout,
1987), and in schools with neighborhood rather than system-wide atten-
dance boundaries (Toles, Schulz, & Rice, 1986). Although these studies
reflect a greater appreciation of the range of factors that contribute to
school dropout, the role of peers in school dropout has not been adequate-
ly considered to date.
A theoretical model of school motivation

In a recent model of school motivation, Ryan and Powelson (1991) argue
that three basic needs are fundamental to learning: feelings of compe-
tence, autonomy, and relatedness. These researchers begin with the as-
sumption that students come to school with a natural and innate curiosity
and interest in learning, which manifests itself as a basic need for feelings
of competence and autonomy as a primary basis for motivation (Ryan,
1991; White, 1963). In their model, competence refers to one's sense of
mastery and accomplishment when challenged optimally (see Harter,
1983; White, 1960). Autonomy refers to the idea of self-regulation or self-
determination (see deCharms, 1968; Deci & Ryan, 1985, 1987). However,
there is another, equally important and natural tendency that is critical to
education - the need to develop connections and associations within a so-
cial matrix, which Ryan and Powelson refer to as a psychological need for
relatedness. Relatedness refers to the interpersonal attachments and bonds
developed between individuals, and is based on a fundamental striving for
contact and alliance with others, enhancing the well-being of all involved.
Ryan (1991) further suggests that the need for relatedness is a social mo-
tive that is evident in one's concern for "what others look for in and from
us" (Ryan, Connell, & Grolnick, 1992, p. 172).
In an historical analysis of the changing context of education, moving
from learning through apprenticeship and participation in adult work, to
the modern day institutions of education (schools), Ryan and Powelson
(1991) suggest that "the institutionalization of education in the modern
era removed the processes of learning and cultural transmission from con-
texts in which children were often guided by adults to whom they were
closely attached and from activities of significance in everyday life"
(p. 49). Despite the obvious benefits of modern education, this transition
created a school context in which children are more isolated from the
adult world, and learn skills that often have no clear or immediate purpose
in their daily lives. Ryan and Powelson also suggest that learning has
evolved into an individualized and/or competitive experience, despite ar-
guments for the efficacy of more cooperative and interpersonal learning
approaches (see Hymel, Zinck, & Ditner, 1993). Such changes have effec-
tively reduced the likelihood that the educational system can maintain stu-
dents' feelings of autonomy and relatedness and, in failing to do so, may
contribute to the academic failure of at least some students.
Thus, Ryan and Powelson (1991) direct our attention to two fundamen-
tal aspects of education beyond the basic goal of fostering learning or aca-
demic competence. The need for autonomy has implications for how
classroom environments are structured (see Eccles et al., 1993). The need
for relatedness has implications for students' interpersonal relationships
with both adults and peers. There is ample evidence to support the notion
that teaching practices that foster student autonomy versus control (the
classroom "climate") are related to greater academic motivation, achieve-
ment, and feelings of competence (e.g., Boggiano & Katz, 1991; Bog-
giano, Main, Flink, Barrett, Silvern, & Katz, 1989; Deci, Schwartz,
Scheinman, & Ryan, 1981; Eccles et al., 1990, 1993; Grolnick & Ryan,
1987; Ryan & Connell, 1989; Ryan & Grolnick, 1986).
Far less empirical support is available for the notion that feelings of
relatedness are important in maximizing student learning, although a
few recent studies suggest that relatedness to parents as well as teachers
is important for school success. Ryan, Stiller, and Lynch (1994) exam-
ined how students' feelings of relatedness to parents, teachers, and peers
were related to school functioning (positive coping, autonomy, perceived
control, and general self-esteem). Results indicated that the quality of re-
ported relatedness to both teachers and parents independently predicted
various aspects of school functioning. No predictive relations were ob-
tained for feelings of relatedness to peers, although Ryan et al. suggest
that this may be due to the fact that peers can exert either a positive or
negative influence on school success. Thus, when the influence of all
peers is considered, an overall positive or negative influence is not clear-
ly evident. Goodenow (1993), however, has demonstrated that feelings of
classroom belonging and peer/teacher support in middle school are sig-
nificant predictors of school motivation and expectations for academic
success.
There has been growing recognition of the importance of peer relations
for school engagement. In discussing early school adjustment, Ladd (per-
sonal communication, 24 Sept., 1994) suggests that peers are the "glue"
that solidifies students' initial involvement in the educational enterprise
(see also Ladd, 1990). The importance of peers does not diminish and may
even increase with age. For example, in our own research we have found
that the transition from elementary to middle school is easier for those
students who report positive peer support, and more difficult for students
who are lonely and dissatisfied with their current peer relations
(McDougall & Hymel, 1995). From interviews with urban high school
students, Firestone and Rosenblum (1988) suggest that school "is where
students can come to be with their friends or where they find activities
other than educational ones to keep them occupied" (p. 10). Andersson
(1994) reports that 70% of Swedish adolescents surveyed indicated that
the "best thing" about school is peers, and more than 40% indicated that if
it were not for their peers, they would not be able to "stand school." Thus,
Andersson suggests that peers may be more important than teachers in
fostering school engagement.
Consistent with the model proposed by Ryan and Powelson (1991), our
view is that student participation and involvement in school (factors that
may protect the student against school dropout) are to a large degree de-
pendent on how much the school environment contributes to students'
feelings of both autonomy and relatedness which, in turn, contribute to
positive academic performance and feelings of competence. Like Ryan
and Powelson, we feel that both parents and teachers play a critical role in
maximizing student feelings of both autonomy and relatedness. However,
although parents and teachers have traditionally been viewed as the prima-
ry agents of children's scholastic motivation (e.g., Blumenfeld, 1992), de-
velopmental and educational psychologists have increasingly recognized
the role of peers in the socialization process (e.g., Asher, 1983; Berndt &
Ladd, 1989; Hartup, 1983; Hartup & Sancilio, 1986; Ladd, 1988; Parker
& Asher, 1987). The influence of peers may be particularly critical during
adolescence, when peers take on added significance in one's life (e.g.,
Berndt & Ladd, 1989; Brown, 1990; Buhrmester, 1990; Claes, 1992;
Csikzentmihalyi & Larson, 1994; Hartup, 1993; Larson & Richards,
1991). Adolescents reportedly spend twice as much time with peers than
with their family (Csikzentmihalyi & Larson, 1984; Larson & Richards,
1991), and peers fulfill a developmental need that cannot be met by par-
ents or other adults (Berndt, 1982; Hartup, 1993; Seltzer, 1982). More-
over, Steinberg, Dornbusch, and Brown (1992) suggest that although par-
ents play a critical role in students' long-term educational plans and goals,
"peers are the most potent influence on their day-to-day behaviors in
school (e.g., how much time they spend on homework, if they enjoy com-
ing to school each day, and how they behave in the classroom)" (p. 727).
Thus, especially during adolescence, peers may be particularly important
in providing an educational context in which the student can feel a sense
of belonging and affiliation (relatedness). Although few studies have di-
rectly examined the impact of feelings of peer relatedness on school func-
tioning, there is a growing body of evidence to suggest that various as-
pects of peer relationships are important in determining the likelihood of
school dropout.
The influence of peers on school dropout

Studies have demonstrated consistent links between social competence
and various academic outcomes, suggesting that children who are poorly
accepted and/or aggressive are at greater risk for academic failure than
their more popular or prosocial peers (e.g., Dishion, 1990; Green, Fore-
hand, Beck, & Vosk, 1980; Lambert, 1972; Vosk, Forehand, Parker, &
Rickard, 1982). Given evidence (reviewed above) that academic difficul-
ties are a significant predictor of later school dropout, these findings lead
to the question of whether social factors are themselves directly related to
the likelihood of school dropout. Although limited, there is a growing
body of evidence to suggest that peer relationships do indeed play a criti-
cal role. In our review, we distinguish four aspects of peer influence: prior
social acceptance within the peer group, social isolation versus involve-
ment, the negative influence of peers, and aggressive and antisocial be-
havior.
Prior social acceptance within the peer group

The importance of good peer relations has been most convincingly
demonstrated in studies showing that children who are rejected by peers
are at greater risk for a variety adjustment difficulties in adolescent and
adult life (see Parker & Asher, 1987, for a review). Of interest here is
whether early peer difficulties, as assessed by indices of peer accep-
tance/rejection during the elementary years, place the student at greater
risk for later school dropout. Several older studies have demonstrated di-
rect associations between early social status and later dropout (Barclay,
1966; Gronlund & Holmlund, 1958; Kuhlen & Collister, 1952; Lambert,
1972; Ullmann, 1957). For example, Gronlund and Holmlund (1958)
found that 54% of low accepted boys and 35% of low accepted girls sub-
sequently dropped out of school, as compared with 19% and 4% of more
accepted boys and girls, respectively. Similarly, Barclay (1966) reported
that students who were not well accepted by peers were two to three times
more likely later to drop out of school than their more accepted class-
mates. In more recent studies, using sociometric measures that allow for
the distinction of various social status subgroups (e.g., rejected, neglected,
controversial, popular, average), there is also evidence to suggest that chil-
dren who are rejected during the elementary years are at greater risk for
later school dropout. Specifically, Ollendick et al. (1992) found that 18%
of the children classified as sociometrically rejected (generally disliked)
in grade four had dropped out of school by grade nine, as compared with
5% of popular students, 6% of average status students, 3% of neglected
students, and 9% of controversial status students. Given evidence that stu-
dents are more likely to drop out in the later grades of high school (Center
for Education Statistics, 1993; Statistics Canada, 1993), rejected children
may be even more likely to drop out of school than suggested by the 01-
lendick et al. results. Further, Ollendick et al. found that, relative to popu-
lar peers, rejected students were also more likely to later exhibit problems
of adolescent delinquency, conduct disturbance and substance abuse, to
obtain lower achievement test scores in grade eight, and more grade reten-
tions by grade nine, all factors which have been shown to predict school
dropout (e.g., Barrington & Hendricks, 1989).
Not all studies, however, have demonstrated significant links between
early social status and later dropout. Following 112 fifth grade students
through high school, Kupersmidt and Coie (1990) reported no significant
differences in later dropout rates as a function of sociometric status (using
a categorical modeling analysis). However, consistent with earlier studies,
they reported that about 30% of rejected fifth graders later dropped out,
compared to 20% of average status students, 7% of popular students, and
0% of students who were neglected in grade five. Moreover, Kupersmidt
and Coie provide a footnote indicating that results of a logistic regression
analysis, conducted with only a White subsample, revealed that social
preference among grade five peers (the degree to which they are liked
rather than disliked) did emerge as a significant predictor of later dropout.
Using a different assessment procedure, Cairns et al. (1989) also evaluat-
ed whether early social difficulties predicted later dropout. Although so-
cial status has typically been evaluated using peer assessments, Cairns et
al. used teacher reports to assess popularity in 475 seventh grade students
and followed them through grade eleven. Peer assessments were used to
assess social isolation. In this study, neither popularity nor social isolation
was found to predict subsequent dropout. The failure to demonstrate a re-
lation between social status and later dropout may be attributable to the
use of teacher rather than peer assessments of popularity, or to the exami-
nation of social status during middle, rather than elementary school.
Although findings are somewhat mixed, the majority of these studies do
suggest a link between early problems in peer relations and later school
withdrawal. As well, in a small-scale meta-analysis of the seven available
studies examining peer status (popularity, acceptance, rejection) and later
school dropout, Comfort and Kishor (1994) reported a significant mean ef-
fect size of r = . 19. Although the overall effect size is not large, it does sug-
gest that status among peers plays a role in subsequent school dropout. Sev-
eral questions remain, however, regarding the mechanisms underlying the
relations between early rejection and later dropout. Peer rejection has gen-
erally been shown to be a rather stable phenomenon from one year to the
next (Coie & Dodge, 1983), but such data do not suggest that children who
are rejected within the elementary school classroom are destined to be re-
jected throughout their academic years. Thus, it is not appropriate to as-
sume that early rejection leads to continued rejection, which is itself pre-
dictive of school dropout. Clearly, more extensive longitudinal research on
the stability of peer rejection across the elementary to high school years is
needed, as is research on whether concurrent peer rejection is related to
school dropout. To our knowledge, such data do not presently exist.
Recent research on the functions of peer group formation and peer cul-
tures during adolescence may offer, in part, an hypothesis regarding the
social trajectories of rejected children. As Erikson (1959) and others (e.g.,
Marcia, 1980) have noted, one of the major developmental tasks of ado-
lescence is to develop a sense of identity - adolescents must determine
who they are, what they are all about, and what they are going to do in the
future. As students move from the smaller, homogeneous classrooms of
the elementary school to the larger social world of the secondary school,
the number and variety of peers and cliques available to them increases
substantially. In their search for identity, then; adolescents are able to
choose from a wider array of possibilities than was afforded within the el-
ementary context. These diverse opportunities for affiliation may provide
a chance to overcome the early rejection experienced by some students
during the elementary years. Support for this contention comes from a
qualitative study by Kinney (1993) who examined the everyday experi-
ences of adolescents whose high school peers had labeled them as unpop-
ular during their earlier school years. Through extensive interviews, Kin-
ney identified a pattern whereby students who were previously labeled as
unpopular "nerds" in middle school were able to obtain a more positive
perception of themselves as "normal" and accepted by at least some peers
within the more differentiated social context of the high school. Kinney
suggested that one of the mechanisms responsible for this transformation
from "nerd" to "normal" was the quest for identity formation - a need to
find a peer group niche with whom one can identify. Adolescents seemed
to experience the transition from middle to high school as a time of
increased opportunities to find a peer group niche, recalling that such op-
portunities "provided alternative domains to achieving school-wide popu-
larity in which students could feel adequate and successful" (p. 30). Thus,
the transition to high school may provide some students with an opportu-
nity to rid themselves of the stigma of peer rejection and find a peer group
which offers support for school-related goals.
Other research on peer cultures and adolescent subgroups, however,
suggests that the opposite effect can take place as well - that the peer
group with whom one comes to identify can advocate a value system that
does not support school involvement and participation. For example, Eck-
ert (1989), in her work on social categories in high schools, has identified
a subgroup of "burnouts" - students who were alienated from the social
and academic functioning of school. Such subgroups often fostered the
development of "anti-school" feelings and a dislike for all aspects of
school, such as rules, teachers, and school activities. Thus, the shift to the
larger social world of secondary school may afford students an opportuni-
ty to change their social status, but the effects of such a shift may vary de-
pending on the value system supported by the newly-found social niche, a
point to which we will return shortly. Future research may benefit from
consideration of whether dropout rates vary across peer subgroups within
the high school community.
In summary, it is not clear whether peer rejection constitutes a highly
stable phenomenon that persists through the elementary years and into
adolescence, directly contributing to the likelihood of school dropout, or
whether early peer rejection is merely a precursor to other forms of disen-
gagement from school, serving as one of many factors that together make
school an undesirable context for some students. Consistent with the latter
hypothesis, Kupersmidt, Coie, and Dodge (1990), in reviewing research
on the prediction of school dropout from indices of earlier social status,
conclude that "social rejection might play a unique, incremental role in
this prediction. One reason for this may be that being rejected by the peer
group may make coming to school an aversive experience for adolescents
and thus motivate them to think of leaving school" (p. 290). Still, several
authors suggest that perceived social isolation or peer rejection per se
leads to school dropout (e.g., Pittman, 1986; Valverde, 1987). In the next
section, we review evidence concerning the relation between perceived
social isolation (as opposed to actual peer rejection) and the likelihood of
school dropout.
Social isolation versus involvement

Although the preceding review suggests that early peer rejection is predic-
tive of later school dropout, not all rejected children drop out of school,
and some well-accepted students also drop out. One possibility is that
only those students who perceive themselves to be rejected or isolated
within the peer group are at risk for dropout. Research on children's social
self-perceptions suggests that actual and perceived peer rejection may not
be synonymous. For example, studies have shown that rejected children
are significantly more likely to report feelings of loneliness and social dis-
satisfaction than their more accepted peers, but there is also considerable
variability in children's social self-perceptions, especially among rejected
children (Asher, Parkhurst, Hymel, & Williams, 1990). Perhaps only those
students who acknowledge their social difficulties and who feel isolated
are more likely to dropout, regardless of actual levels of peer rejection.
Many have suggested that students at risk for dropout experience feelings
of alienation, disenfranchisement, or isolation (see Finn, 1989), yet sur-
prisingly few empirical studies have directly examined perceptions of so-
cial isolation among dropouts.
As an initial means of evaluating the degree to which feelings of social
isolation play a role in the decision to drop out of school, we can examine
the reasons students offer for their decision to leave school early. When
asked about their reasons for leaving, a substantial number of dropouts
emphasize peer-related factors, although generalizations are difficult, giv-
en the variable ways in which reasons for dropping out have been catego-
rized across studies. For example, results of some studies (Pittman, 1986;
Rumberger, 1983) suggest that perceived rejection by peers is not neces-
sarily a major reason for school dropout, as far as reports by dropouts
themselves are concerned, although problems with peers may well have
been subsumed within other categories such as "disliked school" (Rum-
berger, 1983) or "unhappy school experience" (Pittman, 1986). Gastright
(1987) reported that only 3% of dropouts indicated that "other students"
were a reason for dropping out. In contrast, in a recent U.S. sample (Cen-
ter for Education Statistics, 1993), 14.5% of dropouts indicated that not
getting along with other students was a reason for dropping out, and
24.2% of dropouts reported that feeling like they did not belong was a
contributing factor in their decision to drop out. Thus, when open-ended
questions are used to assess reasons for school dropout, with researchers
free to categorize responses in their own way, the degree to which students
cite interpersonal difficulties as a major factor in their dropout decision is
difficult to ascertain, although some reports suggest that peer difficulties
may be a factor for a number of dropouts.
Discrepant findings are also evident when more direct evaluation
methods are used. When dropouts were asked to indicate the degree to
which they agreed with various reasons for dropping out, Tidwell (1988)
found that 20.5% agreed that "students at the school" were a reason for
leaving school. In contrast, McCaul (1989), using a similar methodology,
reported that only 6.3% of dropouts agreed that not getting along with stu-
dents was a reason for their dropping out.
Results of several retrospective studies also provide some support for
the notion that school dropouts are more likely to feel socially isolated or
alienated within the high school. Seidel and Vaughn (1991) found that
Learning Disabled (LD) dropouts reported greater social alienation from
peers than did LD nondropouts, although the generalizability of these
findings to non-LD students is questionable. Dohn (1991) found that Dan-
ish dropouts reported feeling more excluded by their classmates than did
nondropouts. In an inner-city sample, Fagan and Pabon (1990) observed
that dropouts were more socially isolated than graduates, as indicated by
self-reported immersion in friendship networks combined with reported
participation in social activities. In a qualitative study comparing Hispan-
ic dropouts versus graduates, Valverde (1987) found that dropouts report-
ed more feelings of alienation/rejection and fewer friendships than did
graduates (see also Williams, 1987). Although results of these studies sug-
gest a relation between feelings of social isolation and school dropout, the
retrospective nature of these studies makes it difficult to determine
whether these self-perceptions existed before the decision to leave school,
or whether they emerged in a post hoc fashion as a rationalization for the
decision to leave school early.
Other studies have examined whether students' feelings of isolation,
alienation, or lack of belonging predict school dropout using experi-
menter-developed, self-report instruments (e.g., Pittman, 1991). Informa-
tion on the psychometric properties of these measures is not always re-
ported, and the conceptual underpinnings of these scales are not always
clear. For example, Wehlage and Rutter (1986) reported that dropouts feel
alienated and rejected in school, but their measure of alienation was em-
bedded within a larger set of items, including perceived teacher interest
and discipline practices. Elliott and Voss (1974) reported that students
who would later drop out reported greater feelings of social isolation, but
the majority of items in their assessment reflected school participation
and extracurricular involvement rather than feelings of isolation per se.
Although a lack of school participation may stem from feelings of isola-
tion (or vice versa), it may also be the result of other factors, such as job-
or family-related demands.
Several authors have argued that involvement in school and extracur-
ricular activities constitutes an index of social integration within the

school context (e.g., Kelly & Pink, 1972; Tinto, 1975). Although some-
what distinct from feelings of social isolation as discussed here, there is
ample evidence to suggest that dropouts are significantly less likely to
participate in school-based extra-curricular activities and/or sports than
are students who do not drop out (Ekstrom et al., 1986; Elliott & Voss,
1974; Hinojosa & Miller, 1984; Kelly & Pink, 1972; Pittman, 1991;
Smith, Tseng, & Mink, 1971; Statistics Canada, 1993; Thomas, 1954;
Walters & Kranzler, 1970). Whether this lack of participation in school
activities is a cause, a consequence, or even a correlate of social isolation
is unclear, but these findings do suggest that the dropout is less involved
in the social aspects of the high school.
In summary, despite suggestions that dropouts may feel socially isolat-
ed, there is little unequivocal empirical support for this notion within the
current dropout literature. For example, when we examine the sponta-
neous reasons students give for leaving school early, some studies indicate
no clear relations between early school leaving and reported peer difficul-
ties (e.g., Pittman, 1986; Rumberger, 1983), whereas other studies show
that a substantial number of dropouts (24.2%) report feeling that they did
not belong as a factor contributing to their decision to leave school early
(Center for Education Statistics, 1993). Results of several retrospective
studies consistently demonstrate that feelings of social isolation are more
likely among dropouts than among graduates, but the retrospective nature
of these findings makes it difficult to determine whether perceived social
isolation is an antecedent of school dropout or a rationalization which
emerges after the fact, owing perhaps to a need to cut ties with one's
school friends after leaving school. There is considerable evidence indi-
cating that dropouts are less likely to participate in extracurricular activi-
ties and/or sports within the school context, suggesting that, at least to
some extent, dropouts are not well-integrated or involved in the social ac-
tivities afforded by schools. However, it is not clear whether a lack of so-
cial involvement stems from feelings of social isolation or from other de-
mands such as work or family. More extensive research is needed,
particularly prospective, longitudinal studies, before definitive conclu-
sions can be reached.
In the absence of clear evidence that perceived social isolation con-
tributes to the likelihood of school dropout, we consider a somewhat dif-
ferent hypothesis regarding the social relations that characterize the early
school leaver. That is, the high school dropout may in fact experience a
reasonable degree of social integration and may enjoy a number of estab-
lished peer relationships, but with peers who do not identify with or par-
ticipate in the school context and who do not encourage school comple-
tion. These arguments are consistent with Finn's (1989) "participation-
identification model" of school dropout, which suggests that students who
drop out are less likely to be active participants in school and classroom
activities, and thus fail to develop a feeling of identification with or
"bonding" to school. Consistent with Ryan and Powelson's (1991) notion
of relatedness, Finn reviews evidence to suggest that students typically de-
velop an internalized feeling of belonging within the school. Two distinct
components are critical here: belonging and valuing. Students must feel a
part of the social world of the school and must value educational success.
As indicated above, research has demonstrated a rather consistent relation
between lack of participation in school and subsequent school dropout.
Thus, dropouts do not appear to be engaged in or committed to the social
activities that typify most educational institutions. At the same time, many
dropouts evidently do not necessarily feel socially isolated, suggesting
that they are involved in some form of satisfying social networks. As Finn
suggests, the social affiliations that characterize early school leavers may
not be ones that enhance feelings of belonging or identification with
school and may not foster the maintenance of positive school values. Re-
search relevant to these hypotheses is reviewed in the next section.
The negative influence of peers

If most school dropouts are not the socially isolated individuals that some
suggest them to be, what is the nature of their social affiliations during the
high school years? Some have suggested that potential dropouts tend to
affiliate with others who are also at risk for early school leaving, and to-
gether disengage from school (e.g., Mensch & Kandel, 1988; Pittman,
1991). Consistent with this hypothesis, Cairns et al. (1989) found that stu-
dents who would eventually drop out tended to affiliate with other stu-
dents who were also at risk for school dropout. Thus, although school
dropouts may not be socially isolated, they may be victims of negative
peer influence. There is some evidence to support these speculations.
First, results of a recent national survey of school dropouts (Statistics
Canada, 1993) indicated that students who leave school early are more
likely to have friends outside of school than are students who graduate. In
addition, most graduates (80%) reported having friends who believed that
high school completion was very important, and only a few graduates
(2%) reported having friends who felt that completing high school was
unimportant. In contrast, fewer than half of the dropouts surveyed report-

ed having friends who believed completing high school was important,
and 12% of dropouts reported that their friends felt high school comple-
tion was not very important. Similar findings have emerged from qualita-
tive studies of school dropout (Delgado-Gaitan, 1986; Farrell, Peguero,
Lindsey, & White, 1988), suggesting that involvement in peer groups that
embrace a noneducational orientation can contribute to the decision to
drop out of school early. Research has also shown that students who have
friends who like school (Hinojosa & Miller, 1984), who get good grades
and are interested in school or attend classes regularly (Ekstrom et al.,
1986) are more likely to graduate from high school. Using a path analysis,
Pittman (1991) provides evidence to suggest that of all the variables he ex-
amined in the High School and Beyond data set, peers' interest in school
had the largest impact on students' own interest in school which, in turn,
had the largest direct influence on later school dropout. Other research in-
dicates that dropouts tend to have lower educational and occupational ex-
pectations than do graduates (Wehlage & Rutter, 1986), and that they per-
ceive their friends to have lower educational aspirations as well (Ekstrom
et al., 1986; Rumberger, 1983). Results of retrospective studies have also
shown that dropouts were significantly more likely to have friends who
had dropped out of high school than were graduates (Alpert & Dunham,
1986; Dunham & Alpert, 1987), although only a small proportion of
dropouts (2.5%, McCaul, 1989; 8%, U.S. Center for Education Statistics,
1993) acknowledge that one of the reasons that they dropped out was that
their friends were dropping out. It may be that students who are at high
risk for dropping out of school simply share a number of negative school
values with like-minded friends who are also at risk for school dropout or
who have already left school. These findings underscore the importance
of determining the role that peers play in the socialization of educational
values.
To understand the roles that peers might play, we must first understand
the general developmental trajectory of academic motivation across the
adolescent years. Results of several studies suggest a general increase in
negative attitudes toward learning and achievement (Eccles, Midgley, &
Adler, 1984) and a gradual decline in academic motivation over the early
adolescent years (see Eccles & Midgley, 1989, 1990, for reviews). In par-
ticular, Harter (1981) has shown that, as students move through our educa-
tional system, there is a gradual shift from an intrinsic interest in learning
during the elementary years to a more extrinsic motivational orientation
by adolescence. These general declines do not characterize all adoles-
cents, however, with a large proportion of students exhibiting no major

change in motivation and some showing increases in intrinsic motivation
(Harter, Whitesell, & Kowalski, 1986, as cited in Harter, 1992).
There is some evidence to suggest that academic motivation may be
socialized within students' self-selected peer networks. For example, stud-
ies have shown that students tend to associate with classmates who exhib-
it a similar level of school performance (Tesser, Campbell, & Smith,
1984) and/or a similar motivational orientation toward school (Kinder-
mann, 1993). Such self-selected peer affiliations may set the stage for en-
hancing or decreasing students' value for school. At least two studies have
examined whether students' motivational orientation can be modified by
peer group experiences. In a small TV study, Kindermann (1993) demon-
strated that the motivational orientation of children's peer groups was pre-
served over a school year (despite considerable changes in group member-
ship), and that the motivational orientation of the group influenced
individual changes in motivation across the year. Berndt, Laychak, and
Park (1990) demonstrated that discussions with friends about achieve-
ment-related dilemmas can increase similarity in achievement-related de-
cisions, especially if the friendship is characterized by greater cooperation
and acceptance, and less aggression and conflict.
More recent models of academic motivation have moved from consid-
eration of two contrasting motivational orientations (intrinsic versus ex-
trinsic) to a third orientation: internalized motivation (Harter, 1992, this
volume; Ryan et al., 1992). According to Harter (1992), internalized moti-
vation represents the effects of socialization within the culture, such that
academic behaviors that are initially controlled by extrinsic rewards grad-
ually become internalized, as the student comes to view these behaviors
and attitudes, valued by significant others, as important for the self. Al-
though internalized motivation is not the idealized intrinsic motivation
that characterizes the very young student's natural curiosity and interest in
learning, it does represent a self-directed or self-regulated motivational
orientation, one that operates in absence of extrinsic rewards. Our view is
that internalized motivation is particularly likely to be influenced by sig-
nificant others in one's life, most notably, parents, teachers, and peers. As
noted earlier, research has shown that students who drop out of school are
more likely to associate with peers who do not value educational excel-
lence or do not view high school graduation as important (e.g., Cairns et
al., 1989; Delgado-Gaitan, 1986; Farrell et al., 1988; Statistics Canada,
1993), and who are perceived to have lower educational aspirations (Ek-
strom et al., 1986; Rumberger, 1983). Thus, students at risk for dropout
may be more likely to affiliate with peers who do not positively influence
the internalization of culturally prescribed values, or who fail to con-
tribute positively to the socialization of internalized motivation.
Given arguments that an internalized motivational orientation is social-
ized by significant others in one's life, it is also important to remember
that students who drop out tend to come from families that provide less
encouragement for educational success (Ekstrom et al., 1986; Howell &
Frese, 1982) and have parents who themselves exhibit lower levels of edu-
cational attainment (Lloyd, 1978; Statistics Canada, 1993). Thus, the in-
fluence of both parents and peers on the development of an internalized
motivational orientation may be less likely for the dropout. Steinberg et al.
(1992) specifically suggest that both parenting practices and peer support
for academic attainment play critical roles in determining school out-
comes. Although Steinberg et al. did not address the issue of school
dropout per se, they concluded that "youngsters whose friends and parents
both support achievement perform better than those who receive support
only from one source but not the other, who in turn perform better than
those who receive no support from either" (p. 727). They further suggest
that the relative availability of parental and peer support for academic at-
tainment varies across ethnic groups, with the "congruence of parent and
peer support [being] greater for White and Asian-American youngsters
than for African-American and Hispanic adolescents" (p. 727). Thus,
Steinberg et al. conclude that parents and peers can offset or compensate
for the influence of each other on school values.
In summary, results of several studies indicate that dropouts are more
likely to affiliate with peers outside of school, with peers who have them-
selves dropped out, and/or with peers who are also at risk for dropout.
Further, dropouts tend to associate more with peers who evidence less
positive educational values, interest, or motivation, and who have lower
educational aspirations. It may be that early peer rejection (during the ele-
mentary years) constitutes the beginning of a cycle in which rejected stu-
dents who are not well integrated within the mainstream classroom begin
to associate with similar others - those who do not support the "school
game" - and together these students gradually disengage from school. On
the basis of these findings, we suggest that the peer group plays a critical
role in the socialization of educational values and academic motivation, a
relationship that to date has been relatively unrecognized within the litera-
ture. Although parents and teachers have traditionally been seen as the pri-
mary agents of socialization with regard to school motivation, our review
suggests that students' academic motivation and perceived school value
are also influenced by those peers with whom they are closely associated,
and this influence can be either positive or negative. The peer group may
also play an important role in the socialization of aggressive and antiso-
cial behavior, a characteristic that has also been linked to the likelihood of
school dropout, as discussed in the next section.
Aggression and antisocial behavior

Considerable evidence exists regarding the relationship between aggres-
sive behavior and school dropout. Although aggressive behavior cannot be
considered an aspect of peer influence per se, it does reflect students' dif-
ficulty in getting along with peers and therefore is considered relevant to
the present review. Studies comparing dropouts with graduates consistent-
ly demonstrate that dropouts are rated by both teachers and peers as ex-
hibiting more aggressive behavior than graduates (Kuhlen & Collister,
1952; Lambert, 1972; see Parker & Asher, 1987, for a review). Moreover,
ratings of aggressive behavior during elementary school (Ensminger &
Slusarcick, 1992; Kupersmidt & Coie, 1990) as well as early adolescence
(Cairns et al., 1989) have been shown to be a strong predictor of subse-
quent school dropout, even after the influence of other variables such as
gender, race, peer rejection, excessive absences, and low grades (Kuper-
smidt & Coie, 1990), or race, SES, grade retention, maturation, and popu-
larity (Cairns et al., 1989) are taken into account. As is the case for indices
of early peer rejection, however, the studies reviewed herein have been
concerned with early aggressive behavior as a predictor of subsequent
dropout, and have not addressed the concurrent relations between aggres-
sive or antisocial behavior and school dropout. In addition to aggressive
behavior, school dropouts report more regular alcohol and drug use than
do graduates, and also are more likely to have criminal records than are
graduates (Ekstrom et al., 1986; Elliott & Voss, 1974; Mensch & Kandel,
1988; Steinberg, Blinde, & Chan, 1984; Statistics Canada, 1993).
Finn (1989) suggests that these negative social behaviors may be a re-
action to the frustration some students feel within the school context.
Specifically, in his "frustration-self-esteem model" of school dropout,
Finn suggests a cycle that begins with poor school performance (attribut-
able to student academic deficits and/or poor instruction), which leads to
low self-esteem on the part of the student in response to the frustration
and embarrassment caused by school failure. These feelings of personal
failure fuel reactive opposition to the school context, which is held re-
sponsible, observed in the form of various kinds of disruptive or opposi-
tional behavior (aggression, delinquency, classroom disruption, skipping

class). According to Finn, some versions of this model suggest that, as the
resulting oppositional behavior becomes more disruptive with age, school
personnel direct greater attention toward the behavior problems them-
selves, with less and less attention directed toward the learning difficulties
that underlie them. As a result, the student falls further behind academi-
cally, and behavior problems escalate, becoming increasingly entrenched.
Other versions of this model suggest that low self-esteem leads to a search
for alternate activities (such as oppositional behavior) in which the student
might feel some form of success. Even though such activities and behav-
iors are less socially sanctioned by the schools, they may garner the ap-
proval of similar nonparticipant or antisocial peers (see Elliott & Voss,
1974; Finn, 1989 for relevant supportive literature). In the end, the frus-
trated student is at increased risk for school dropout. According to this
model, then the negative and antisocial behavior that characterizes some
dropouts emerges as a response to academic frustration, but may be main-
tained and supported through associations with like-minded peers. Thus,
selective peer affiliations may contribute to the socialization of antisocial,
aggressive behavior in the same way as they contribute to the socialization
of negative school values.
Relative and cumulative effects

In light of evidence reviewed thus far, it seems clear that academic, famil-
ial, school, and social variables may all play a role in the decision to leave
school early. Although the relative contribution of these factors has not
been determined in research to date, we may be better able to understand
the underpinnings of school dropout if we consider the process to be a
gradual, sequential, or cumulative one, resulting from multiple contribut-
ing factors (e.g., Farrell, 1990; Finn, 1989). Yet few studies examining the
relative contribution of various factors in predicting school dropout have
included consideration of social factors, as reviewed below.
The relative impact of social and academic factors has been considered
in a model of dropout from higher education proposed by Tinto (1975)
and recently applied to the phenomenon of high school dropout (e.g.,
Pittman, 1991). According to Tinto, the decision to drop out is largely in-
fluenced by two constructs: academic integration and social integration.
Academic integration refers to the degree to which a student identifies
with "those environmental features that are part of the school's academic
mission" (Pittman, 1991, p. 209), as reflected in such things as grades,
teacher evaluations, etc.; whereas social integration refers to the "student's

identity within the school's social structure" (Pittman, 1991, p. 209), as re-
flected by social involvement, participation, integration, etc. Applying
Tinto's framework to higher education in a college setting, Pascarella and
colleagues (e.g., Pascarella & Chapman, 1983) have demonstrated that
these two factors exhibit what they term "compensatory relationships" in
predicting subsequent dropout. Thus, academic difficulties may be more
critical to the decision to drop out for students who are not well integrated
socially, whereas social difficulties may be more critical for students who
are not doing well academically.
Despite the appeal of Tinto's model, few researchers have attempted to
explore the relative or interactive influence of both social and academic
variables on school dropout within the high school setting. However, re-
sults of these studies do support the idea that social and academic factors
may interact in important ways in predicting the likelihood of school
dropout. Kelly and Pink (1972) considered the combined effect of grade
point average and participation in extracurricular activities on early school
withdrawal. Although no statistical analyses were presented, results did
suggest a combined effect, with reported dropout rates of 2% for students
with both high grades and high social participation, 5% for students with
high grades and low social participation, 22% for students with high so-
cial participation and low grades, and 38% for students with both low
grades and low social participation. More recent studies by Cairns et al.
(1989) and Kupersmidt and Coie (1990) have used logistic regression
analyses to examine the relative contribution of academic and social vari-
ables to later dropout, and both found aggressive behavior, in combination
with academic indicators, to be a significant predictor of dropout. Specif-
ically, Kupersmidt and Coie found that the probability of dropout was
lowest (9.8%) for students who were neither excessively absent from
school nor aggressive, and higher for students who were either only exces-
sively absent from school (27%) or only aggressive (45%), but highest
(73.7%) for students who were both excessively absent and highly aggres-
sive. Similarly, Cairns et al. found that students who were viewed by
teachers as exhibiting low aggressiveness and average to high academic
competence were least likely to drop out of school (7.5%), whereas stu-
dents who were viewed by teachers as exhibiting both high aggressiveness
and low academic competence were the most likely to dropout (64%). In
between these extremes were students who were viewed by teachers as
low in academic competence but not aggressive (22% dropped out), and
students who were viewed as average to high in academic competence but
highly aggressive (31.3% dropped out). Taken together, results of these

studies underscore the value of exploring the relative and combined influ-
ence of social and academic variables on the decision to drop out of
school. Future research would benefit from consideration of the indepen-
dent and overlapping contributions of a wider variety of social as well as
academic variables, as well as the relative influence of familial and other
variables in this regard.
It may also be important to consider the relative contribution of social
factors and characteristics of the school environment to early school with-
drawal. As noted, academic motivation tends to decline over the early ado-
lescent years (Eccles & Midgley, 1989, 1990), with a gradual shift from
an intrinsic to a more extrinsic motivational orientation by adolescence
(Harter, 1981), although not all students show such a decline (Harter,
Whitesell, & Kowalski, 1986, as cited in Harter, 1992). Harter (1992) re-
views evidence to suggest that these changes in motivational orientation
are related to changes in students' self-perceptions of their ability (with
students who viewed themselves as increasing in academic competence
showing corresponding increases in academic motivation), as well as to
perceived changes in the educational environment. In particular, the shift
toward a more extrinsic and less intrinsic motivational orientation was as-
sociated with perceived changes in educational practices from the elemen-
tary to the high school setting, including greater emphasis on academic
performance, competence evaluation, and social comparison, all of which
served to highlight variations in ability. With regard to the present empha-
sis on peer influence, we suggest that the perceived increase in the
salience of social comparison among students, coupled with the generally
competitive nature of the educational enterprise, creates a situation in
which classmates are increasingly viewed as adversaries rather than part-
ners in learning. Thus, characteristics of the school environment that fos-
ter competition rather than cooperation among students may serve to dis-
enfranchise some students from the larger social milieu of the school and
may inadvertently reduce the positive influences of peer support for aca-
demic success.
Conclusions
The decision to drop out of school appears to involve an interplay of aca-
demic, familial, school, and social factors. The influence of these factors
may be additive, with each contributing to the likelihood of dropout, or
may vary across individuals, with particular factors being more critical for
some students than others. Moreover, the decision to drop out of school
does not occur overnight, but likely reflects the cumulative impact of a va-
riety of experiences over an extended period of time. In the present chap-
ter, we have attempted to highlight the role of peers in this process, argu-
ing that peer influences have been relatively neglected within the dropout
literature to date. Based on arguments put forward by Ryan and Powelson
(1991) that feelings of relatedness are critical to academic functioning and
school success, we propose that feelings of relatedness with peers may be
particularly important during adolescence, and that peers may play a criti-
cal role in encouraging students to stay in school or drop out. We do not
wish to dismiss the importance of academic, familial, and school factors,
which has been clearly documented in the literature extant. Notwithstand-
ing, the purpose of the present review was to highlight yet another factor
in the decision to drop out or stay in school - the role of peers.
In support of our arguments, we reviewed research demonstrating that
various aspects of students' peer relationships are related to school
dropout. First, it seems that early peer rejection during the elementary
school years places students at greater risk for later dropout, although the
mechanisms underlying this association are unclear. It may be simply that
students who are rejected do not feel a sense of relatedness with their
peers and, as a result, disengage from school and drop out. However, there
are no data at present to suggest that dropouts are rejected at the time that
they decide to leave school, although future research should examine this
possibility. Nor are there clear data to suggest that dropouts feel socially
isolated (although there is suggestion that some of them do). Thus, the hy-
pothesis that continued peer rejection (or perceived isolation) itself consti-
tutes a major factor in the decision to drop out remains a question for fu-
ture research.
Given evidence that early social difficulties are to some extent related
to academic problems, it may be that only socially rejected students who
experience academic failure are at increased risk for dropout. We suggest
that both academic and social integration are critical for school success.
Although few studies have directly examined the "compensatory relation-
ships" that may exist between social and academic integration, results of
these studies show that students who experience both social and academic
failure are at considerably higher risk for later dropout. Thus, future re-
search should attempt a more in-depth examination of the combined ef-
fects of academic and social problems on the likelihood of school dropout,
as well as the additional influence of familial and school variables.
Other research reviewed in this chapter, however, suggests that the in-
fluence of early peer rejection on subsequent dropout may be more com-

plex. Based on the preceding review, it appears that a more plausible sug-
gestion is that the student who experiences rejection during the early school
years may be at greater risk for developing associations during adolescence
with like-minded classmates who fail to socialize positive academic values,
who discourage active school participation, and who may actively support
or encourage deviant behavior, all of which are factors known to be associ-
ated with school dropout. Although this proposed developmental trajectory
must be considered speculative at present, the available supportive evi-
dence suggests that further research on this possibility is warranted.
In considering the influence of peers on the decision to leave school
early, it is important to recognize that there may be multiple pathways that
students can traverse as they negotiate their way to the completion of high
school. Although a high school education seems necessary for survival in
today's society, for some students the decision to leave school early may
be entirely appropriate for their circumstances and may ultimately repre-
sent a positive life choice. For most students, however, the decision to
drop out of school may be the lamentable result of a number of factors,
one of which concerns involvement with peers. Given the research re-
viewed herein, several different pathways can be suggested. For some stu-
dents, early peer rejection may be one of a number of factors that together
contribute to students' feelings of alienation and isolation within the
school context, and eventually lead them to view school as an undesirable
place to stay. As Finn (1989) suggests in his "participation-identification"
model of school dropout, such students fail to participate actively in
school and classroom activities and fail to "bond" to the school enterprise,
increasing the likelihood of dropping out. For other students, early peer
rejection, coupled with academic difficulties, may lead to a reduced sense
of self-esteem, which in turn leads to increased oppositional behavior, as
suggested in Finn's "frustration-self-esteem" model of school dropout. To-
gether, these factors may enhance the likelihood of identification and af-
filiation with other aggressive and antisocial students who devalue the im-
portance of school generally and the completion of a high school
education in particular. Similarly, early peer rejection and aggression may
also mark the beginning of a cycle in which students begin to affiliate
with other at-risk students who together disengage from school. These ex-
amples represent only three of a potentially larger number of pathways
that may lead a student to drop out of school. If the additional influences
of familial and school variables are also considered, the possible trajecto-
ries become even more numerous.
The present review suggests that the likelihood of school dropout de-
pends on a number of factors, some of which reflect characteristics of the
student (e.g., academic and familial factors, as implied in an individual
deficit model of dropout) and others that reflect characteristics of the so-
cial context of the school. We live in a social world, and adolescence
marks a period during which the triadic interplay of the individual, the
family and the peer group becomes particularly salient. Accordingly, the
decision to leave school is perhaps most appropriately viewed, not as the
result of deficits in the individual, but as an interaction of student charac-
teristics (academic competence, motivation), familial background and
support, and the influence of the peer group with whom one interacts on a
daily basis. It becomes imperative, then, that future research consider the
relative and compensatory contributions of each of these, in terms of both
identifying the causes of school dropout and developing preventive efforts
to keep students in school. As one example, we are currently involved in
an ongoing partnership with an inner-city high school to explore the mul-
tiple factors associated with school dropout. In this prospective, longitudi-
nal study, we began by identifying students who are perceived to be at
high risk for school dropout according to teacher-, peer-, and self-reports,
and are examining whether these students differ on a variety of factors in-
cluding academic performance, motivation, familial characteristics, as
well as current peer rejection, social isolation, school involvement, and
aggressive behavior. As the entire sample is followed over time, we hope
to examine the complex interplay of individual and social variables as
they predict actual dropout.
Our hope is that by highlighting the influence of peers in determining
student motivation to stay in school, we will stimulate further research on
the relative impact of multiple factors, including peers, on school dropout.
Although our review of this literature has focused primarily on the poten-
tially negative influence of peers, there is another perspective that also
must be recognized. Just as peers can encourage early school leaving, or
fail to foster positive school values and motivation, it is equally possible
for peers to positively influence students to value school and to complete
their education. For example, Rebane and Schonert-Reichl (1994) have
developed a unique classroom for grade nine students identified by the
school staff as being at high risk for subsequent dropout. In this "Bridge"
class-room, the teacher has used the principles of a "just community"
(e.g., Kohlberg & Higgins, 1987) to develop a democratic classroom
aimed at developing a sense of community and collective solidarity
among students and between the students and the teacher. Interviews with
the students at the end of the first year suggest that this experimental
classroom was effective in developing more positive attitudes toward
school among these students. The students attributed such changes to sev-
eral differences they perceived between the "Bridge" classroom and the
regular school program, including positive peer and teacher-student rela-
tionships, opportunities for democratic decision-making, and opportuni-
ties to negotiate their own academic curriculum. This experimental
"democratic" classroom had a marked influence on improving school at-
tendance, especially in the "Bridge" classes, which may be a first step to-
ward reducing dropout risk. In the words of one student, "I just hope that
I'll still be in this class next year. If it wasn't for Bridge class, I wouldn't
be in school." (Rebane & Schonert-Reichl, 1994, p. 3). Consistent with
Ryan and Powelson (1991), our hope is that by fostering the development
of both autonomy and relatedness, we can improve students' sense of
school belonging. And this sense of community and belonging may, for
some students, be a key factor in the decision to stay in school. As we be-
gin to appreciate the impact of peers on school dropout, we may be better
able to maximize the positive rather than negative influence of peers, es-
pecially with regard to school engagement.
Acknowledgment
Preparation of this chapter was supported in part by a Faculty of Educa-
tion Partnership Research Grant from the University of British Columbia,
in conjunction with the British Columbia Ministry of Education to authors
Schonert-Reichl and Hymel.
References
Alpert, G., & Dunham, R. (1986). "Keeping academically marginal youths in
school" A prediction model. Youth and Society, 17, 346-361.
Andersson, B. (1994, February). School as a setting for development - a Swedish
example. Paper presented at the biennial meeting of the Society for Research on
Adolescence, San Diego, CA.
Asher, S. R. (1983). Social competence and peer status: Recent advances and na-
ture directions. Child Development, 54, 1427-1434.
Asher, S. R., Parkhurst, J. T., Hymel, S., & Williams, G. A. (1990). Peer rejection
and loneliness in childhood. In S. R. Asher and J. D. Coie (Eds.), Peer rejection
in childhood (pp. 253-273). NY: Cambridge University Press.
Barclay, J. R. (1966). Sociometric choices and teacher ratings of school dropout.
Journal of Social Psychology, 4, 40^5.
Barrington, B. L., & Hendricks, B. (1989). Differentiating characteristics of high

school graduates, dropouts, and nongraduates. Journal of Educational Re-
search, 82, 309-319.
Berndt, T. J. (1982). The features and effects of friendship in early adolescence.
Berndt, T. J., & Ladd, G. W. (1989). Peer relationships in child development. NY:
Wiley.
Berndt, T. J., Laychak, A. E., & Park, K. (1990). Friend's influence on adoles-
Biemiller, A., & Meichenbaum, D. (1993). How to make destreaming work: To-
wards successful integration. Unpublished manuscript, University of Toronto.
Blumenfeld, P. C. (1992). Classroom learning and motivation: Clarifying and ex-
panding goal theory. Journal of Educational Psychology, 84, 272-281.
Boggiano, A. K., & Katz, P. (1991). Maladaptive achievement patterns in stu-
dents: The role of teachers' controlling strategies. Journal of Social Issues, 47,
35-51.
Boggiano, A. K., Main, D. S., Flink, C, Barrett, M, Silvern, L., & Katz, P. A.
(1989). A model of achievement in children: The role of controlling strategies
in helplessness and affect. In R. Schwarzer, H. V van der Ploeg, & C. D. Spiel-
berger (Eds.), Advances in test anxiety research (pp. 13-26). Lisse, Nether-
lands: Swets & Zeitlinger.
Brown, B. B. (1990). Peer groups and peer cultures. In S. S. Feldman & G. R. El-
liott (Eds.), At the threshold: The developing adolescent (pp. 171-196). Cam-
bridge, MA: Harvard University Press.
Buhrmester, D. (1990). Friendship, interpersonal competence, and adjustment in
preadolescence and adolescence. Child Development, 61, 1101-1111.
Cairns, R. B., Cairns, B. D., & Neckerman, J. J. (1989). Early school dropout:
Configurations and determinants. Child Development, 60, 1437-1452.
Catterall, J. S. (1985). On the social costs of dropping out of school (CERAS Re-
port No. 86-SEP-3). Stanford, CA: Stanford University, School of Education.
Catterall, J. S. (1987). An intensive group counseling dropout prevention interven-
tion: Some cautions on isolating at-risk adolescents within high schools.Ameri-
can Educational Research Journal, 24, 521-540.
Center for Education Statistics (September, 1993). Dropout Rates in the United
States: 1992, U.S. Department of Education, Office of Educational Research
and Improvement.
Claes, M. E. (1992). Friendship and personal adjustment during adolescence.
Journal of Adolescence, 15, 39-55.
Coie, J. D., & Dodge, K. A. (1983). Continuities and changes in children's social
status: A five-year longitudinal study. Merrill-Palmer Quarterly, 29, 261-282.
Combs, J., & Cooley, W. W. (1968). Dropouts: In high school and after school.
American Educational Research Journal, 5, 343-363.
Comfort, C, & Kishor, N. (1994, June). Relations of self-esteem, alienation and

popularity to school dropout: A meta-analysis. Paper presented at the annual
meeting of the Canadian Society for the Study of Education. Calgary, Alberta.
Csikzentmihalyi, M., & Larson, R. (1984). Being adolescent. NY: Basic Books.
Crespo, M, & Michelena, J. (1981). Streaming, absenteeism and dropping-out.
Canadian Journal of Education, 6, 40-55.
deCharms, R. (1968). Personal causation: The internal affective determinants of
behavior. NY: Academic.
Deci, E. L., & Ryan, R. M. (1985). Intrinsic motivation and self determination in
human behavior. NY: Plenum.
Deci, E. L., & Ryan, R. M. (1987). The support of autonomy and the control of
behavior. Journal of Personality and Social Psychology, 53, 1024-1037.
Deci, E. L., Schwartz, A. J., Sheinman, L., & Ryan, R. M. (1981). An instrument
to assess adults' orientations toward control versus autonomy with children:
Reflections on intrinsic motivation and perceived competence. Journal of Edu-
cational Psychology 74, 642-650.
Delgado-Gaitan, C. (1986). Adolescent peer influence and differential school per-
formance. Jo urnal of Adolescent Research, 1, 449-462.
Dishion, T. J. (1990). The family ecology of boys' peer relations in middle child-
hood. Child Development, 61, 874-892.
Dohn, H. (1991). "Drop-out" in the Danish high school (gymnasium): An investi-
gation of psychological, sociological and pedagogical factors. International Re-
view of Education, 37, 415-428.
Dryfoos, J. G. (1990). Adolescents at risk. NY: Oxford University Press.
Dunham R. G., & Alpert, G. P. (1987). Keeping juvenile delinquents in school: A
prediction model. Adolescence, 22, 45-57.
Eccles, J. S., McCarthy, K. A., Lord, S. E., Harold, R., Wigfield, A., & Aberbach,
A. (1990, April). The relationship of family factors to self-esteem and teacher-
rated adjustmentfollowing the transition to junior high school environment. Pa-
per presented at the annual meeting of the Society for Research on Adoles-
cence, Atlanta, GA.
Eccles, J. S., & Midgley, C. (1989). Stage/environment fit: Developmentally ap-
propriate classrooms for early adolescents. In R. Ames & C. Ames (Eds.) Re-
search on Motivation in Education (vol. 3). San Diego, CA: Academic.
Eccles, J. S., & Midgley, C. (1990). Changes in academic motivation and self-per-
ception during early adolescence. In R. Montemayor, G. Adams, & T. P. Gullot-
ta (Eds.), From Childhood to Adolescence: A transitional period? (pp.
134-155). Newbury Park: Sage.
Eccles, J. S., Midgley, C, & Adler, T. F. (1984). Grade-related changes in the
school environment: Effects on achievement motivation. In J. G. Nicholls (Ed.),
The development of achievement motivation (pp. 283-331). Greenwich, CT:
JAI.
Eccles, J. S., Midgley, C, Wigfield, A., Buchanan, C. M, Reuman, D., Flanagan,
C, & Maclver, D. (1993). Development during adolescence: The impact of

stage-environment fit on young adolescents' experiences in schools and in fam-
ilies. American Psychologist, 48, 90-101.
Eckert, P. (1989). Jocks and burnouts: Social categories and identity in the high
school. NY: Teachers College Press, Columbia University.
Ekstrom, R. B., Goertz, M. E., Pollack, J. M., & Rock, D. A. (1986). Who drops
out of high school and why? Findings from a national study. Teachers College
Record, 87, 356-373.
Elliott, D. S., & Voss, H. L. (1974). Delinquency and dropout. Lexington, MA: D.
C. Heath.
Employment and Immigration Canada (1990). A National Stay-in-School Initia-
tive. Ottawa, ONT: Minister of Supply and Services.
Ensminger, M. E., & Slusarcick, A. L. (1992). Paths to high school graduation or
dropout: A longitudinal study of a first-grade cohort. Sociology of Education,
(55,95-113.
Erikson, E. H. (1959). Identity and the life cycle (revised edition). NY: W. W. Nor-
ton.
Fagan, J., & Pabon, E. (1990). Contributions of delinquency and substance use to
school dropout among inner-city youth. Youth and Society, 21, 306-354.
Farrell, E. (1990). Hanging in and dropping out: Voices ofat-risk high school stu-
dents. NY: Teachers College Press.
Farrell, E., Peguero, G., Lindsey, R., & White, R. (1988). "Giving voice to high
school students" Pressure and boredom, Ya no what I'm saying'? American Ed-
ucational Research Journal, 25, 489-502.
Fine, M. (1991). Framing dropouts: Notes on the politics of an urban public high
school. Albany, NY: State University of New York Press.
Finn, J. D. (1989). Withdrawing from school. Review ofEducational Research, 59,
117-142.
Firestone, W. A., & Rosenblum, S. (1987, April). First year project: A study of
alienation and commitment in five urban districts. Paper presented at the annu-
al meeting of the American Educational Research Association, Washington, D.
C.
Gastright, J. F. (1987). Profile of students at-risk. (ERIC Document Reproduction
Service, No. 313 465).
Goertz, M. E., Ekstrom, R. B., & Rock, D. (1991). Dropouts, high school: Issues
of race and sex. In R. M. Lerner, & J. Brooks-Gunn (Eds.), Encyclopedia of
adolescence, (vol. 1). NY: Garland.
Goodenow, C. (1993). Classroom belonging among early adolescent students: Re-
lationships to motivation and achievement. Journal of Early Adolescence, 13,
21-43.
Green, K. D., Forehand, R., Beck, S. I, & Vosk, B. (1980). An assessment of the
relationships among measures of children's social competence and children's
academic achievement. Child Development, 51, 1149-1156.
Grolnick, W. S., & Ryan, R. M. (1987). Autonomy in children's learning: An ex-

perimental and individual difference investigation. Journal of Personality and
Gronlund, N., & Holmlund, W. (1958). The value of elementary school sociomet-
ric status scores for predicting pupils' adjustment in high school. Education Ad-
ministration Supervision, 44, 225-260.
Hahn, A. (1987). Reaching out to America's dropouts: What to do? Phi Delta
Kappan, Dec, 256-263.
Harter, S. (1981). A new self-report scale on intrinsic versus extrinsic orientation
in the classroom: Motivational and informational components. Developmental
Psychology, 77,300-312.
Harter, S. (1983). Developmental perspectives on the self-system. In E. M. Het-
herington (Ed.), Handbook of child psychology, vol. 4, Socialization, personali-
ty and social development, (4th edition, pp. 275-386). NY: Wiley.
Harter, S. (1992). The relationship between perceived competence, affect, and
motivational orientation within the classroom: Processes and patterns of
change. In A. K. Boggiano & T. W Pitman (Eds.), Achievement and motivation:
A social-developmental perspective (pp. 77-114). NY: Cambridge.
Hartup, W W. (1983). Peer relations. In E. M. Hetherington (Ed.), P. H. Mussen
(Series Ed.), Handbook of child psychology, vol. 4: Socialization, personality,
and social development (pp. 103-196). NY: Wiley.
Hartup, W W (1993). Adolescents and their friends. In B. Laursen (Ed.), Close
friendships in adolescence (pp. 3-22). San Francisco, CA: Jossey-Bass.
Hartup, W. W, & Sancilio, M. F. (1986). Children's friendships. In E. Schopler &
G. B. Mesibov (Eds.), Social behavior in autism (pp. 61-79). NY: Plenum.
Hinojosa, D., & Miller, L. (1984). Grade level attainment among migrant farm
workers in south Texas. Journal of Education Research, 77, 346-350.
Howell, F. M., & Frese, W. (1982). Early transition into adult roles: Some an-
tecedents and outcomes. American Educational Research Journal, 19, 51—73.
Hymel, S., Zinck, B., & Ditner, E. (1993). Cooperation versus competition in the
classroom. Exceptionality Education Canada, 3, 103—128.
Kelly, D. H., & Pink, W. T. (1972). Academic failure, social involvement, and high
school dropout. Youth and Society, 4, 47-59.
Kinderman, T. A. (1993). Natural peer groups as contexts for individual develop-
ment: The case of children's motivation in school, Developmental Psychology,
29, 970-977.
Kinney, D. A. (1993). From nerds to normals: The recovery of identity among ado-
lescents from middle school to high school. Sociology ofEducation, 66, 21-40.
Kohlberg, L., & Higgins, A. (1987). School democracy and social interaction. In
W. M. Kurtines & J. L. Gewirtz (Eds.), Moral development through social in-
teraction (pp. 102-128). NY: Wiley.
Kuhlen, R., & Collister, E. G. (1952). Sociometric status of sixth and ninth-
graders who fail to finish high school. Educational and Psychological Mea-
surement, 12, 632-637.
Kupersmidt, J. B., & Coie, J. D. (1990). Preadolescent per status, aggression, and
school adjustment as predictors of externalizing problems in adolescence.
Peer rejection in childhood (pp. 274-308). NY: Cambridge.
Ladd, G. W. (1988). Friendship patterns and peer status during early and middle
childhood. Journal of Developmental and Behavioral Pediatrics, 9, 229-238.
Ladd, G. W. (1990). Having friends, keeping friends, making friends, and being
ment? Child Development, 61, 1081-1100.
Larson, R., & Richards, M. (1991). Daily companionship in late childhood and
early adolescence: Changing developmental contexts. Child Development, 62,
284-300.
Lambert, N. A. (1972). Intellectual and nonintellectual predictors of high school
status. Journal of Scholastic Psychology, 6, 247—259.
Lloyd, D. N. (1978). Prediction of school failure from third-grade data. Educa-
tional and Psychological Measurement, 38, 1193-1200.
Marcia, J. E. (1980). Identity and adolescence. In J. Adelson (Ed.), Handbook of
Adolescent Psychology (pp. 145-160). NY: Wiley.
McCaul, E. (1989). Rural public school dropouts: Findings from High School and
Beyond. Research in Rural Education, 6, 19-24.
McCaul, E. J., Donaldson, G. A., Coadarci, T., & Davis, W. E. (1992). Conse-
quences of dropping out of school: Findings from High School and Beyond.
Journal of Educational Research, 85, 198-207.
McDougall, P., & Hymel, S. (1995, April). The transition to middle school: The
voice of the consumer Paper presented at the annual meeting of the American
Educational Research Association, San Francisco, CA.
Mensch. B. S., & Kandel, D. B. (1988). Dropping out of high school and drug in-
volvement. Sociology of Education, 61, 95-113.
Ollendick, T. H., Weist, M. D., Borden, M. C, & Greene, R. W. (1992). Sociomet-
ric status and academic, behavioral, and psychological adjustment: A five-year
longitudinal study. Journal of Consulting and Clinical Psychology, 60, 80—87.
O'Sullivan, R. G. (1990). Validating a method to identify at-risk middle school
students for participation in a dropout prevention program. Journal of Early
Adolescence, 10, 209-220.
Pascarella, E. T., & Chapman, D. W. (1983). A multi-institutional, path analytic
validation of Tinto's model of college withdrawal. American Educational Re-
search Journal, 20, 87-102.
Pittman, R. B. (1986). Importance of personal, social factors as potential means
for reducing high school dropout rate. The High School Journal, 7-13.
Pittman, R. B. (1991). Social factors, enrollment in vocational/technical courses,
and high school dropout rates. Journal of Educational Research, 84, 288-295.
Pittman, R. B., & Haughwout, P. (1987). Influence of high school size on dropout
rate. Educational Evaluation and Policy Analysis, 9, 337-343.
Rebane, K., & Schonert-Reichl, K. A. (1994, November). Fostering a moral com-
munity among "at-risk" secondary students: Implication for promoting school
success. Paper presented at the annual meeting of the Association for Moral
Education, Banff, Alberta.
Rumberger, R. W. (1983). Dropping out of high school: The influence of race,
sex and family background. American Educational Research Journal, 20,
199-220.
Rumberger, R. W. (1987). High school dropouts: A review of issues and evidence.
Review of Educational Research, 57, 101—121.
Ryan, R. M. (1991). The nature of the self in autonomy and relatedness. In G. R.
Goethals & J. Strauss (Eds.), Multidisciplinary perspectives on the self (pp.
208-238). NY: Springer-Verlag.
Ryan, R. M, Connell, J. P., & Grolnick, W. S. (1992). When achievement is not
intrinsically motivated: A theory of internalization and self-regulation in
school. In A. K. Boggiano & T. S. Pittman (Eds.), Achievement and motivation:
A social-developmental perspective (pp. 167-188). NY: Cambridge.
Ryan, R. M., & Grolnick, W. S. (1986). Origins and pawns in the classroom: Self-
report and projective assessments of individual differences in children's percep-
tions. Journal of Personality and Social Psychology, 50, 550-558.
Ryan, R. M., & Powelson, C. L. (1991). Autonomy and relatedness as fundamen-
tal to motivation and education, Journal ofExperimental Education, 60, 49-66.
Ryan, R. M., Stiller, J. D., & Lynch, J. H. (1994). Representations of relationships
to teachers, parents and friends as predictors of academic motivation and self-
esteem. Journal of Early Adolescence, 14, 226-249.
Seidel, J. E, & Vaughn, S. (1991). Social alienation and the learning disabled
school dropout. Learning Disabilities Research and Practice, 6, 152-157.
Seltzer, V C. (1982). Adolescent social development: Dynamic functional interac-
tion. Lexington, MA: Lexington Books.
Smith, J. E., Tseng, M. S., & Mink, O. G. (1971). Prediction of school dropouts in
Appalachia: Validation of a dropout scale. Measurement and Evaluation in
Guidance, 5, 31—37.
Statistics Canada (September, 1993), Leaving School, Report for Human Re-
sources and Labour Canada.
Steinberg, L., Blinde, P. L., & Chan, K. S. (1984). Dropping out among language
minority youth. Review of Educational Research, 54, 113-132.
Steinberg, L., Dornbusch, S. M., & Brown, B. B. (1992). Ethnic differences in
adolescent achievement: An ecological perspective. American Psychologist, 47,
723-729.
Stoup, A. L., & Robins, L. N. (1972). Elementary school predictors of high school
dropout among black males. Sociology of Education, 45, 212-222.
chology, 46, 561-574.
Thomas, R. J. (1954). An empirical study of high school drop-outs in regard to ten
possibly related factors. Journal of Educational Sociology, 28, 11-18.
Tidwell, R. (1988). Dropouts speak out: Qualitative data on early school depar-
tures. Adolescence, 23, 939-954.
Tinto, V (1975). Dropout from higher education: A theoretical synthesis of recent
research. Review of Educational Research, 45, 89-125.
Toles, R., Schultz, E. M., & Rice, W. K. (1986). A study of variation in dropout
rates attributable to effects of high schools. Metropolitan Education, 1, 30-38.
Ullmann, C. A. (1957). Teachers, peers and tests as predictors of adjustment.
Valverde, S. (1987). A comparative study of Hispanic high school dropouts and
graduates: Why do some leave school early and some finish? Education and
Urban Society, 19, 320-329.
Vosk, B., Forehand, R., Parker, J. B., & Rickard, K. (1982). A multimethod com-
parison of popular and unpopular children. Developmental Psychology, 18,
571-575.
Walters, H. E., & Kranzler, G. D. (1970). Early identification of the school
dropout. The School Counselor, 18, 97-104.
Wehlage, G. G., & Rutter, R. A. (1986). Dropping out: How much do schools con-
tribute to the problem? Teachers College Record, 374-392.
White, R. W (1960). Competence and the psychosexual stages of development. In
M. R. Jones (Ed.), Nebraska Symposium on Motivation (Vol. 8, pp. 97-141).
Lincoln, Nebraska: University of Nebraska Press.
White, R. W. (1963). Ego and reality in psychoanalytic theory. NY: International
Universities Press.
Williams, S. B. (1987). A comparative study of Black dropouts and Black high
school graduates in an urban public school system. Education and Urban Soci-
ety, 19, 311-319.
Zaslow, M. J., & Takanishi, R. (1993). Priorities for research on adolescent devel-
opment. American Psychologist, 48, 185-192.
14 What's "emotional" about social
motivation? A comment
Sandra Graham
Reading the chapters for this section of Social motivation: Understanding

children s school adjustment reminded me of an incident that occurred
about ten years ago when my older son was beginning middle school.
With some apprehension, I was sitting in the audience at the parent orien-
tation meeting, wondering whether I had made the right decision in allow-
ing him to attend such an academically competitive private school. The
headmaster approached the podium, scanned the sea of nervous-looking
faces, smiled, and then said: "Relax, parents. We know that there are only
two questions your son or daughter will be asking this year: "Where's my
pencil and who's my best friend?".
The insights of this headmaster reflect a common theme underlying the
five chapters in this section of Juvonen and Wentzel's edited volume. That
theme is that the academic outcomes of children and adolescents are inte-
grally related to their social outcomes. Thus, the motivational psycholo-
gist concerned with individual achievement strivings also needs to exam-
ine how social relationships and affiliative concerns influence these
strivings.
Although the specific outcome examined and scope of the coverage
are quite disparate in the various chapters, each in some way is concerned
with social influences on individual motivation. Beginning when children
first enter school, Birch and Ladd (Chapter 9) describe the young child's
relationship with his or her teacher as a determinant of academic adjust-
ment. Relationships characterized by intimacy, nondependency, and the
absence of conflict appear to facilitate the young child's adaptation to the
academic environment, particularly during the transition from kinder-
garten to elementary school.
Adopting a goal theory perspective, Wentzel (Chapter 10) considers
how prosocial goals (e.g., being cooperative and helpful towards others)
and social responsibility goals (e.g., following rules and keeping commit-
346
What s "emotional" about social motivation ? 347
ments) can be predictors of positive academic outcomes. Students who

endorse being kind to others and complying with classroom norms as im-
portant pursuits appear to do better in school and to be more accepted by
both teachers and peers.
Rather than peer influences in general, Berndt and Keefe (Chapter 11)
focus on dyadic relationships, or close friendships, which are character-
ized by equality and mutual respect. Here the data suggest that students
are influenced by both the attitudes of their close friends (e.g., their valu-
ing of academic achievement) as well as the quality of the relationship
(e.g., the degree to which it is mutually satisfying), and these influences
can have both positive and negative effects on academic adjustment.
In the most empirical of the five chapters, Kindermann, McCollam,
and Gibson (Chapter 12) consider how peer networks can be an an-
tecedent to academic engagement of children and adolescents. Even
though children's choice of workmates (networks) show much fluctuation
over the course of a school year, at least one characteristic of selected
group members (i.e., their motivational orientation) remains quite stable
and is predictive of an individual's level of academic engagement.
Finally, Hymel, Comfort, Schonert-Reichl, and McDougall (Chapter
13) discuss the influence of peers on a more molar adjustment outcome -
school dropout in adolescence. It is evident that children who are rejected
by others during the elementary years are at greater risk for early school
withdrawal. These same youngsters are likely to affiliate with peers who
also have become academically disengaged, thus reinforcing the accept-
ability of early dropout. In sum, all of the chapters to some degree exam-
ine the social context of school adjustment, spanning early childhood to
adolescence, including teacher, peer groups, and close friends as the so-
cial influence, and addressing motivationally-relevant outcomes ranging
from grade-level academic achievement to the decision to complete or
withdraw from high school.
There certainly is a precedent for relating achievement concerns to so-
cial concerns, inasmuch as the historically dominant theories and theorists
of motivation implicitly recognized the importance of relationships with
others as determinants of behavior. For example, Freud stated that work
(achievement) and love (affiliation) are the two major motivations in life
(cf. Weiner, 1992); Maslow (1968) placed the need to belong between sur-
vival needs and self-actualization in his needs hierarchy; and motive theo-
rists include affiliative needs with achievement and power needs as the
most important human motives (see McClelland, 1985). Recognizing this
tradition, Baumeister and Leary (1995) recently reviewed a large body of
literature supporting their hypothesis that the need to belong, or the desire
for interpersonal attachments, is a fundamental human motivation.
While acknowledging social influences, however, the field of motiva-
tion has always been and continues to be dominated by the study of in-
trapsychic processes and individual achievement strivings. How relation-
ships shape and are shaped by individual behavior, which is more properly
the domain of social motivation, has never played a prominent role in the
study of personal motivation. By highlighting broad socialization process-
es, that is, the influences of close friends, peer groups, and teachers on
student academic adjustment, the motivation researchers who authored the
present chapters have moved our field a step closer to a needed integration
of the study of personal motivation and the study of social motivation in
achievement settings.
In my own research on African American youth, I have only recently
come to recognize the importance of this kind of integration between
achievement strivings and social relationships. I began my career by
studying academic motivation in African American youth from an in-
trapsychic perspective. I have always believed that far too many minority
children perform poorly in school not because they lack intellectual com-
petencies or even specific learning skills, but because they have low ex-
pectation, feel hopeless, deny the importance of effort, or give up in the
face of failure. These are prototypical individual motivation concerns. But
I now believe that far too many ethnic minority children also perform
poorly in school because they have few friends, they adhere to an opposi-
tional peer culture, or they elicit anger from teachers and peers. These also
are prototypical motivational concerns, but of a social rather than individ-
ual academic nature. What I now try to do is study individual motivation,
or achievement strivings, and social motivation, in this case aggression,
within the context of a general theoretical framework that applies to both
motivational systems.
What do I mean when I talk about social motivation and its importance
for understanding individual motivation? If we take as a starting point the
present authors' reviews documenting the importance of peers and teach-
ers as socializers of individual achievement strivings, then a social moti-
vational analysis should tell us something about process, or how these so-
cial forces influence what individuals do. For example, one might choose
to study because she expects that those in her peer work group will be an-
gry if she does not "pull her weight." In this case, the anticipated reactions
of others guide the expenditure of effort. Or a popular student might invite
a shy classmate for a playdate because he feels sorry for peers who have
What s ''emotional" about social motivation ? 349
difficulty making friends. Here the emotions directed toward others influ-
ence prosocial responding. These examples intimate that one possible di-
mension of a social motivational analysis might be to examine whether in-
dividual behavior is a response to anticipated emotional reactions of
others versus emotions that are elicited about others.
In the remainder of this commentary, I will elaborate on this and other
distinctions as I attempt to describe what I see as some of the dimensions
of a social motivational analysis. Of course, it is possible that how I con-
ceptualize the "social" in social motivation will be perceived as idiosyn-
cratic. Many of the examples I use to support my analysis are drawn from
attribution theory, which reflects my own bias to view the social world
through an attributional lens. This bias notwithstanding, my goal is to
stimulate the reader to think more about how a social motivational analy-
sis might uncover some common themes that link the chapters in this sec-
tion of Juvonen and Wentzel's volume.
Dimensions of a social motivational analysis

The importance of affect
A first important dimension of social motivation is that it is likely to have
an affective component. Students like or dislike their teachers, they feel
accepted or rejected by their classmates, they are sometimes angry at their
parents, proud of their teammates, embarrassed by public failure, etc. Stu-
dents may also attempt to control the feelings that others have about them
and to manipulate what others think by means of affective displays. Thus,
emotional reactions in social contexts have motivational significance.
Interplay between self-directed and other-directed emotions. Emotions

in social contexts may be distinguished as to whether they are self-direct-
ed or other-directed. For example, a pupil may feel guilty or ashamed fol-
lowing failure; these are feelings directed toward the self. In the same con-
text, a teacher may experience anger or pity toward the same failing
student; these are other-directed feelings. How we feel about ourselves
and how others feel about us are often intertwined and may be guided by
the same set of underlying principles. This interrelationship has clearly
been demonstrated in numerous empirical findings based on attribution
theory. Summarizing this work, Weiner (1995) recently provided a theo-
retical analysis of the relations between the self-directed emotions of guilt
and shame and the other-directed emotions of anger and pity based on the
causal construct of perceived controllability in self and others. These rela-

tions are depicted in Table 14.1.
Consider again a failing student who ascribes her poor outcome to ei-
ther a controllable cause (e.g., lack of effort) or to one that is uncontrol-
lable (e.g., low aptitude). These particular attributions have quite disparate
affective consequences. As Table 14.1 shows, one feels guilty when the
causes of personal failure are due to controllable factors. This includes not
only lack of effort, but also any other causal ascription subject to one's
own volitional influence. Shame, in contrast, is more likely to be experi-
enced when personal failures are due to uncontrollable factors such as low
aptitude.
Now consider this same outcome and causal analysis from the perspec-
tive of the teacher. Anger and pity are other-directed emotions that follow
attributions about controllability and the closely associated judgments
about personal responsibility. Perceiving the student's failure as caused by
controllable factors leads to inferences that the person is responsible for
her outcome. Responsibility implies freedom of choice; the student is free
to try hard or not. Attributions to controllability and inferences of respon-
sibility tend to evoke anger (imagine your feelings about the gifted athlete
who never comes to practice). On the other hand, attributing someone
else's failure to low aptitude elicits perceptions that the outcome was un-
controllable and the person was not responsible. Ascriptions to uncontrol-
lability and inferences of nonresponsibility then often evoke feelings of
pity or sympathy (think of your reaction to the retarded child who contin-
ually experiences academic difficulty).
Emotional reactions not only reflect causal thinking; they also provide
guides for future action. As depicted in Table 14.1, self-directed guilt due
Table 14.1. Self- and other-directed emotions and their behavioral

consequences as a function of causal controllability (adapted from Weiner,
1995, p. 265)
Attribution for failure
Perspective Controllable Uncontrollable

Self-directed
Emotion guilt shame
Behavior enhanced performance diminished performance
Other-directed
Emotion anger pity
Behavior punishment, reprimand help
to lack of effort often functions as a motivational enhancer; it propels the

individual to work harder in the future. In contrast, self-directed shame
due to perceived low aptitude may interfere with achievement strivings.
Shame often leads to withdrawal or giving up in the face of failure. Other-
directed pity and anger (which are the respective complements of shame
and guilt - i.e., they have the same causal antecedents) are linked to
prosocial versus punitive behavior. Pity instigates approach behavior and
the desire to help a needy person, whereas anger begets punishment, repri-
mand, and the withholding of help.
These emotions and behavioral responses of both student and teacher
in response to student failure are likely to be occurring simultaneously
and to be mutually reinforcing. For example, anger from a teacher can be
a cue to the student that he is not trying hard enough (see Graham, 1984).
Assuming that this attributional message is accepted, the student feels
guilty and augments her achievement strivings. By the same token, the
low aptitude student who feels shame and displays achievement withdraw-
al is likely to elicit especially high levels of sympathy from the teacher
and offers of help. Excessive sympathy and unsolicited social support
then further exacerbate the student's perception that "I cannot."
In sum, affective cues from others can influence a variety of thoughts,
including self-attributions. Communicated emotions therefore play an im-
portant role in self-esteem and individual achievement strivings. This kind
of analysis, where emotions play a key role as mediators between causal
thought and action, provides one framework for integrating personal moti-
vation (how we feel about ourselves as a determinant of behavior) and so-
cial motivation (how others feel about us as a determinant of behavior).
Anticipated versus actual emotions of others. The above examples

of emotional reactions in a social context indicate that people respond to
the affective displays of others. Now imagine another situation, one that
many of us have probably encountered in our role as teachers. A student
fails to complete an important term paper on time. Let us further assume
that the reason for this outcome was that she went skiing with friends in-
stead of working on the paper. On the day the assignment is due, the stu-
dent comes to her professor and immediately says: "I was sick over the
week-end and couldn't finish my paper. Can I have an extension?" Provid-
ing an uncontrollable explanation for achievement failure, and assuming
that the explanation is accepted, preempts anger from the teacher. In this
case, then, the "spring" of action is the anticipated rather than actual reac-
tion of another. Actual emotions from others are those that are communi-
cated through facial expressions, verbal messages, and bodily or postural

gestures. Anticipated emotions, in contrast, are possible feeling states that
we expect from others based on our understanding of the norms and rules
of social interaction. We may behave obediently to ward off others' poten-
tial anger, comply with their requests to insure their gratitude, show mod-
esty so as not to arouse their jealousy, etc.
Individuals often respond to the anticipated emotions of others, and the
motives underlying their actions provide important dimensions of a social
motivational analysis. For example, a number of well-documented impres-
sion management strategies from social psychology research are guided
by the desire to control others' emotions. That is, individuals often act to
create a favorable impression of themselves and thus gain social approval
by manipulating how others feel about them (see Snyder & Higgins,
1988).
In attribution research on account giving, we have been able to demon-
strate how excuses, confessions (apologies), and other kinds of accounts
are used by children and adults to alter perceptions of responsibility and
anticipated anger (e.g., Graham, Weiner, & Benesh-Weiner, 1995; Weiner,
Graham, Peter, & Zmuidinas, 1991; also see Juvonen, this volume). For
example, socially-adjusted children as young as age 6 or 7 and even boys
known to have conduct problems recognize that communicating an uncon-
trollable excuse for breaking an appointment (e.g., "I was sick" versus "I
wanted to be with other friends") alters perceptions of responsibility and
preempts others' anger.
Emotions, helping, and aggression. Prosocial behavior, including help-

ing and acceptance, and antisocial behavior, including aggression and re-
jection, are two broad classes of behavior that are a natural context for
studying social motivation. A great deal of research documents that the
proximal determinants of both helping and aggression are often emotions.
Recall the earlier discussion of teacher affective cues. Teachers who think
student failures are caused by uncontrollable causes feel pity and want to
help. Furthermore, a very reliable finding in the attribution literature on
helping behavior is that emotions, more than causal beliefs, are the imme-
diate determinant of helping (see review in Weiner, 1995). Thus, attribu-
tion theorists propose a particular set of relations between thinking, feel-
ing, and acting whereby causal ascriptions about another determine
affects directed toward those others; and affects, in turn, guide behavior.
Emotions therefore provide an important link between thinking and doing.
A similar analysis can be applied to the study of peer aggression. One
robust finding in the peer aggression literature is that aggressive children

overattribute hostile intent to others in ambiguous situations (see Dodge,
1993). That is, in situations where they experience negative outcomes due
to the actions of a peer, such as being pushed while waiting in line, and it
is unclear why this outcome occurred, aggressive children are more likely
to infer that the peer's actions were initiated "on purpose." Attributions to
hostile intent are related to perceptions of controllability and responsibili-
ty and therefore elicit anger. Anger, in turn, is an antecedent to aggression.
Just as in the study of helping behavior, we have been able to document
that emotions, more than causal beliefs, are the direct determinants of ag-
gressive responding (Graham Hudley, & Williams, 1992). Substantively
we interpret our findings to mean that when aggressive children reason
about the causes of their peer provocation, much of the relationship be-
tween what they think (e.g., "He did it on purpose") and the way they in-
tend to behave (e.g., "I am going to get even") can be accounted for how
they feel (e.g., "I'm really angry about this"). Our thoughts tell us how to
feel and our feelings tell us what to do.
We have also found that normally adjusted children use emotions in
this manner to guide their decisions about social acceptance or rejection.
For example, Graham and Hoehn (1995) documented that children as
young as age 5 or 6 differentially reacted to aggressive versus shy/with-
drawn classmates based on perceptions of responsibility, intentionality,
and feelings of pity and anger. In this study, excessively shy hypothetical
classmates were perceived as not responsible for their social maladjust-
ment, and they elicited pity. Aggressive peers, in contrast, were perceived
as responsibly for their behavior, and they evoked anger. Furthermore, in-
dependent of causal beliefs, feelings of sympathy elicited (were directly
predictive of) social acceptance of the shy student, whereas feelings of
anger predicted rejection of the aggressive student. Thoughts guide feel-
ings and feelings guide behavior. That is the motivational sequence that I
believe to be quite prevalent in everyday social interaction at school, in the
classroom and on the playground, and even among very young children.
Summary. The above sections describe how emotions play a key role in
understanding how individual behavior shapes and is shaped by relation-
ships with others (which is my working definition of social motivation).
Three functions of emotion in a social motivational analysis have been
highlighted. First, emotions from others are sources of information about
the self (i.e, our failures are due to low aptitude or lack of effort), and
these self-ascriptions then have implications for individual achievement
strivings. Second, anticipated emotions of others instigate the desire to

control those affective states by means of strategic impression manage-
ment tactics. And third, our beliefs about the causes of others' behavior
determine how we feel about them; and these feeling states, in turn, guide
both prosocial (helping, acceptance) and antisocial (aggression, rejection)
behavior.
Although none of the chapters in this section of Juvonen and Wentzel's
volume is organized around the study of affect, most are implicitly con-
cerned with self-, other- directed, or anticipated emotions. For example,
Berndt and Keefe review research on need for social approval as an impor-
tant motive that helps explain the influence of friends on academic adjust-
ment. Some adolescents may study hard to gain the approval of others (i.e.,
so that their friends will like them), whereas others may avoid academic ef-
fort to achieve acceptance (i.e., so that their friends will not dislike them). I
have already suggested that the need for social approval can sometimes in-
stigate impression management tactics that allow an individual to manage
the emotions of others. Similarly, Kindermann et al.'s study of the social-
ization effects of peer networks seems to implicitly draw on anticipated
emotional reactions of others as motivational determinants. If affiliating
with highly motivated peer networks has a positive effect on student en-
gagement, while associating with disaffected networks has a negative ef-
fect, this intimates that the individual's achievement strivings are at least
partly influenced by the anticipated reactions of the members of those net-
works (e.g., "If I don't do my share, they'll be angry at me"). Hymel et al.'s
discussion of aggression and peer rejection as antecedents of school
dropout fits within an analysis of the motivational significance of emotions
directed toward others. The anger that is aroused when others are perceived
as responsible for their undesirable behavior is known to elicit the very be-
haviors that Hymel et al. present as precursors of early school withdrawal.
Finally, the chapters that focus on child-teacher relationships as an-
tecedents of academic adjustment (Birch & Ladd and Wentzel) in part may
reflect the child's responsiveness to the affective cues of others. Teachers
are powerful sources of information that implicate the self, and much of this
information is likely to be conveyed by affective displays.
The empirical findings and theoretical literatures reviewed in these
chapters are quite amenable to analyses that incorporate emotions from a
social motivational perspective. A systematic mapping of the function of
emotions in interpersonal contexts might be a useful way to organize these
vast and diverse literatures that examine social influences on academic
achievement.
Although the focus of my commentary has been on emotions, I want to

briefly address two other dimensions of social motivation that I believe
uncover other less apparent themes common to the chapters in this section
of Juvonen and Wentzel's edited volume.
The targets of social influence

A second dimension of a social motivational analysis applied to the pre-
sent chapters pertains to the target of social influence. The kinds of social
relationships described in this volume occur at the dyadic level, as in close
friendships that are examined by Berndt and Keefe, or at the group level,
as investigated by Hymel et al., and in the peer network research of Kin-
dermann et al. Relationships may also be egalitarian (i.e., involving
peers), or hierarchical, where the source of influence is one's teacher or
parents (Birch & Ladd; Wentzel, both this volume).
Children may respond to the anticipated or actual reactions of close
friends versus peers versus authority figures like teachers or parents.
However, their goals may differ depending on the target of their reactions.
This was illustrated in a recent study that my colleagues and I conducted
on children and adolescents' use of excuses following a social transgres-
sion (Graham et al., 1995). In this study, third through eighth grade boys
were asked to imagine that they failed to keep an appointment with a
friend or their mother, and the reason for the social transgression was ma-
nipulated to be either controllable or uncontrollable. For each of these
causes of social misconduct, respondents rated the likelihood that they
would reveal that "true" cause. We documented that all children were less
likely to disclose controllable than uncontrollable causes - that is, they
were more likely to substitute an excuse rather than admit personal re-
sponsibility for their social transgression. This is consistent with attribu-
tional analyses of excuses as altering perceptions of responsibility away
from the self.
More pertinent to the present discussion, other data from this study
suggested that the children's goals of account giving may have been differ-
ent when the target was their mother versus a friend. Before presenting the
manipulated causes, we asked respondents to tell us what they would say
to their mother or friend if they failed to keep their appointment because
they decided to "hang out with (other) friends" (i.e., a controllable reason
for social transgression). When the target of the transgression was a peer,
children's free responses indicated that the goal of excuse giving was to
alter peers' perceptions of responsibility in order to maintain the social
bond. Children's answers referred most frequently to concerns about not

hurting the other person's feelings and protecting the friendship. For
mothers, however, the open ended responses made frequent reference to
offering an excuse in order to minimize their parent's anger and reduce the
likelihood that they would be punished for their behavior. Thus, an im-
pression management tactic can be guided by either relationship-enhanc-
ing goals (promoting social acceptance) or self-protective goals (avoiding
punishment from others) depending on who the actor is attempting to in-
fluence.
A similar analysis was reported by Juvonen and Murdock (1993) in a
study of eighth graders' communications to peers, teachers, and parents
about the reasons for their academic failure (also see Juvonen, this vol-
ume). Juvonen and Murdock documented that eighth grade students were
more likely to tell peers that they failed because of lack of effort. This is
consistent with the notion that adolescents often experience a general de-
valuing of effort, so that having one's peer group perceive you as not try-
ing is likely to promote social acceptance. In contrast, these same respon-
dents were more likely to communicate to teachers and parents that they
failed because of low aptitude. Recall that in attributional analyses, per-
ceived low aptitude in others elicits pity, help, and the absence of repri-
mand (see Table 14.1). Thus, adolescents prefer to tell authority figures
such as parents and teachers that they were not responsible for their failure
in order to avoid criticism and punishment.
In sum, relationship-enhancing goals and self-protective goals seem to
be two important social goals that guide much of children's behavior, in-
cluding their tactics of impression management. How these different goals
interact with the targets of social influence is a relatively unexplored topic
in social motivation research.
Proximal versus distal social influence

Yet a third dimension of a social motivational analysis suggested by the
present chapters concerns the conditions of the social influence. One con-
dition may be defined as that in which the actor comes in direct contact
with another person or group of persons. For example, peer networks and
teachers in the classroom both intimate direct contact in the here and now,
where the social presence may provide specific cues to the actor about
how to behave in a particular situation. If someone in my work group
scribbles on my paper and I perceive his behavior as intended, I am likely
to respond with hostile retaliation (see Dodge, 1993). Thus the behavior of
someone else, as well as how that behavior is interpreted, is an immediate
stimulus to act aggressively.
In contrast, there are conditions where others are an antecedent to an
actor's behavior, but the social influence is more distal inasmuch as there
is no immediate contact or specific cues. For example, if I am studying for
an exam rather than going out with friends, and wondering what these
friends might be thinking about me, I am responding to possible reactions
of others with whom I subsequently will come into contact. Their "pres-
ence" is psychological and does not provide specific cues to guide my
current actions, although the influence may still be strong and capable of
initiating and/or intensifying my behavior.
I suspect that the more molar the adjustment outcome examined, the
more likely the social influences will be distal rather than proximal, and
the more difficult it will be to predict their effects. Thus, studying broad
achievement outcomes such as school withdrawal (Hymel et al.) are prob-
ably less amenable to a social motivational analysis. This may partly ex-
plain why, as many of the authors note, it has been difficult to determine
the conditions under which peer groups have positive versus negative ef-
fects on a student's general academic adjustment.
Summary
Table 14.2 summarizes the dimensions of a social motivational analysis as
outlined here and a representative research topic associated with each. The
underlying message of this analysis is that the study of social influences
on school adjustment might benefit from greater attention to (1) the inter-
play between self-directed and other-directed emotions; (2) how children's
strategies for navigating their social lives change with contextual (e.g.,
target) demands; and (3) specific classes of social and achievement be-
havior with predictable antecedents rather than global adjustment out-
comes. Other dimensions could surely be articulated based on a reading of
the chapters in this book, and the choice of these three is not independent
of my own theoretical and empirical biases. Nonetheless, I do believe that
they provide a common language for discussing a number of themes that
permeate all five chapters in this section of Juvonen and Wentzel's vol-
ume.
The work of the authors featured here also suggests other fruitful di-
rections for research guided by a social motivational analysis. For exam-
Table 14.2. Dimensions of a social motivational analysis and illustrative

research topics
Dimension Research topic
I. Emotions How feelings of others influence

Self-directed vs. other-directed feelings about the self
Anticipated vs. actual emotions of others
Emotions as mediators between thoughts and actions
II. Targets of Social Influence Goals of impression management
Friendship dyads (e.g., maintaining the social bond
Peer groups vs. avoiding punishment)
Authority figures
teachers
parents
III. Proximal vs distal influences Predicting general vs. specific
Peers and others present in the here and now adjustment outcomes
General social influences or those about whom
one ruminates
pie, the study of peer influences on academic outcomes naturally impli-

cates the study of achievement values, yet this has been a relatively ne-
glected topic in motivation research. This may partly reflect measurement
difficulty as well as the focus on cognitive (e.g., expectancy) determinants
of achievement strivings. Whereas the study of cognitive expectancy vari-
ables can be carried out at the intrapsychic level and need not involve oth-
ers, it is not possible to study values independent of their social context. I
believe that understanding values and how they get expressed in the
broader context of parent, teacher, and peer influences represents one of
the most promising topics for motivation researchers.
Secondly, I talked a great deal in this commentary about the impor-
tance of affect in social motivation research, but only a relatively cir-
cumscribed set of attributionally-determined emotions was actually dis-
cussed. Affective life in the classroom entails a myriad of social
emotions including, for example, pride, envy, jealousy, gratitude, embar-
rassment, and humiliation, to name but a few. Once emotion researchers
liberated themselves from exclusive reliance on facial expressions and
physiological indices to measure emotion, such feeling states became le-
gitimate topics of study, and many now have growing empirical litera-
tures (see Tangney & Fischer, 1995). Furthermore, when expressed in the
classroom, these affects are integrally linked to social rejection and ac-
ceptance, and the time therefore seems right to explore their motivation-
al significance within the context of social influences on academic ad-

justment.
A final note
There was a time when the study of personal motivation, or the "why" of
achievement strivings, and the study of social motivation, or the "why" of
interpersonal behavior, were independent pursuits. As the chapters in this
volume demonstrate, signs of rapprochement are now abundant. This sig-
nals a new vitality for the field of motivation, as new areas of inquiry are
opened up and new opportunities are found for cross fertilization with
other disciplines.
References
Baumeister, R., & Leary, M. (1995). The need to belong: Desire for interpersonal
attachments as a fundamental human motivation. Psychological Bulletin, 117,
497-529.
Dodge, K. (1993). Social-cognitive mechanisms in the development of conduct
disorder and depression. In L. Porter & M. Rosenzweig (Eds.), Annual review
of psychology (vol. 44, pp. 559-584). Palo Alto, CA.: Annual reviews.
Graham, S. (1984). Communicating sympathy and anger to black and white stu-
dents: The cognitive (attributional) antecedents of affective cues. Journal of
Personality and Social Psychology, 47, 40—54.
Graham, S., & Hoehn, S. (1995). Children's understanding of aggression and
withdrawal as social stigmas: An attributional analysis. Child Development, 66,
1143-1161.
Graham, S., Hudley, C, & Williams, E. (1992). Attributional and emotional deter-
minants of aggression in African American and Latino early adolescents. De-
velopmental Psychology, 28, 731—740.
Graham, S., Weiner, B., & Benesh-Weiner, M. (1995). An attributional analysis of
the development of excuse giving in aggressive and nonaggressive African
American boys. Developmental Psychology, 31, 374-284.
Juvonen, J., & Murdock, T. B. (1993). Hoe to promote social approval: The effect
of audience and outcome on publically communicated attributions. Journal of
educational Psychology, 85, 365-376.
McClelland, D. (1985). Human motivation. Glenville, 111.: Scott, Foresman & Co.
Maslow, A. (1943). A theory of human motivation. Psychological Review, 50,
370-396.
Snyder, C, & Higgins, R. (1988). Excuses: Their effective role in the negotiation
of reality. Psychological Bulletin, 104, 23-35.
Tangney, J., & Fischer, K. (Eds.) (1995). Self-conscious emotions: The psychology
of shame, guilt, embarrassment, and pride. New York: Guilford Press.
Weiner, B. (1992). Human motivation: Metaphors, theories, and research. New-
bury Park, CA.: Sage.
Weiner, B. (1995). Judgments of responsibility: A foundation for a theory of social
conduct. New York: The Guilford Press.
Weiner, B., Graham, S., Peter, O., & Zmuidinas, M. (1991). Public confession and
forgiveness. Journal of Personality, 59, 281-312.
Author index
Aber, J. L., 280 Ball, S. J., 254, 256, 265

Aberbach, A., 318 Baltes, P. B., 279
Adams, R. G., 284 Bandura, A., 108, 110, 149, 154-164, 168,
Adan, A. M., 262 170-171, 174, 176, 226, 229, 252, 254
Adler, T. F., 12, 13, 15, 226, 279, 328 Barclay, J. R., 320
Ageton, S. S., 72 Barker, G. P., 58, 279
Ainley, M. D., 280 Barrett, M , 318
Ainsworth, M. D. S., 208 Barrington, B. L., 314, 321
Allen, J. D., 230 Bartholomew, K., 31
Alpert, G. P., 328 Batson,C. D., 127, 129, 136
Altham, P. M. E., 71 Batson,J. G., 129, 136
American Association for University Women Baumeister, R. F., 3, 49, 347
(AAUW), 35 Baumrind, D., 234
Ames, C , 2, 4, 57, 60, 188-189, 279 Beauvais, F., 88
Ames, R. E., 2, 279 Beck, A. T., 74
Amrikhan, J., 45, 49 Beck, S.J., 206, 281, 320
Andersson, B., 319 Belfiore, P. J., 158
Andrews, K., 255 Belmont, M. J., 183, 209, 212, 235, 279, 281,
Ansley, L., 250 283,295-296
Anson, A. R., 266 Benesh-Weiner, M., 47, 352, 355
Aristotle, 260 Bennett, R. E., 283
Arkin, R. M., 49 Benning,J.J.,219,228
Armacost, R. L., 66 Bergin, C. A. C , 129
Aronson, E., 8 Berk, L. E., 167
Asarnow, J. R., 70, 212 Berkowitz, L., 103, 116
Asendorpf, J. B., 82 Berkowitz, S. D., 284, 287-288
Asher, S. R., 59, 66, 70-72, 75, 78, 80, 82-84, Berndt, T. J., 49, 53, 60, 71-72, 77-78,
87,98, 102-105, 107, 109-112, 117, 88-89, 98, 175, 203, 209, 248-249,
119, 120-121, 181-187, 191,203-206, 251-252, 256-259, 261-262, 264-267,
209, 214-215, 228, 260-261, 263,266, 269-270,281,291,329
280-281,284,319-320,324,331 Bersoff, D. M., 270
Ayers, M., 80-81 Berzonsky, M D., 60
Ayers-Nachamkin, B., 45 Biemiller, A., 313
Bierman, K. L., 72, 78, 118
Bailey, D. B., 292 Birch, S. H., 201, 203, 206-207, 209-212,
Bailey, T., 163 214,218
Bakeman, R., 287 Bishop, J.H., 248, 250-251
361
362 Author index
Blackstone, T., 268 Capaldi, D. M, 280, 303

Blinde, P. L., 330-331 Caparulo, B. K., 259
Block,! H., 116 Carleton-Ford, M. A., 13
Block, P., 74 Carlson, C. L.,217
Blumenfeld, P. C.,235,319 Cassidy, J., 59, 206
Blyth, D. A., 13,57,60 Catterall,J. S.,313
Boggiano, A. K., 259, 279, 318 Cauce, A. M., 270
Boivin, M, 260 Causey, D., 249, 261-262, 266
BoldizarJ.P.,33,82, 107,110,191 Center for Education Statistics, 313-316, 321,
Bond, R. N., 83, 86 324, 326, 328
Bonner, S., 155 Cerullo, F., 283
Borden, M. C, 69, 320 Cervone, D., 168, 171
Bossert, S. T., 235 Chaiken, S., 252, 256, 258
Bouffard-Bouchard, T., 162 Chan, K. S., 330-331
Bowlby, J., 73, 206, 209-210, 213-214 Chao, C. C, 269
Braaten, D. O., 49 Chapman, D. W., 333
Brakke,N. P., 69,212,216 Chapman, M, 53, 127, 129, 279
Brand, S., 262 Chase, C, 136
Brasic, J. R., 85 Chen, C, 280
Breiger, R. L., 285 Cheng, C. H., 258
Brett, J. F., 290 Cicchetti, D., 208
Bricker, W., 88 Claes, M. E.,319
Britton, B. K., 173 Clark, K. E., 201
Brophy, J., 279, 283 Clark, M. L., 80-81
Brown, B. B., 54, 57, 60, 78, 241, 251, 253, Clasen, D. R.,251,253,258
258,319,330 Coadarci, T., 313
Brown, M. M., 120 Cobb, J. A., 228
BrunerJ. S., 175 Cobbs,G.,31,37
Brush, U 13 Cody, M. J., 49
Bryan, J. H., 175 Cohen, A. K., 281, 289-291, 302
Buchanan, C. M, 318 Cohen, E.G., 174,266-267
Buhrmester, D., 68, 70, 73,76-78, 203, 209, Coie, J. D., 66, 69-70, 82, 86, 98-99, 106,
236, 238, 248, 260, 264, 281, 319 118,183, 205-207, 212-213,216-218,
Bukowski, W. M., 70, 80, 202-203, 259, 281, 228, 253, 266-267, 280-281, 303,
284,303 321-323,331,333
Burchinal, M. R., 72, 292 Coleman, C. C, 203-204
Burleson, B. R., 201 Coleman, J. S., 54, 248-251
Busceme, S., 268 Collins, W. A., 264
Bussey, K., 110 Collister, E.G., 320,331
Butler, L. J., 47, 60 Colton, M. E., 66
Combs, J., 314
Comfort, C, 321
Cain, K. M., 186, 192 Condry, J., 12
Cairns, B. D., 71, 282, 285, 289-290, 314, Connell, J. P., 4, 12, 15, 131,201,215,
321,327,329,331,333 229-230, 279-283, 295, 317-318,329
Cairns, R. B., 71, 282, 284-285, 287, Cook, T. D., 266
289-291, 314, 321, 327, 329, 331, 333 Cooley, C. H., 24-25
Calkins, S. D., 82 Cooley, W. W.,314
Cameron, A. M., 279 Cooper, D. M, 290
Campbell, J., 71,256,281,329 Cooperman, G. Y., 129
Cantor, N., 131 Coplan, R. J., 82
Cantrell, V. L.,217 Coppotelli, H., 69, 205, 281, 303
Author index 363
Corno, L., 280 Dunham, R. G., 328

Corsini, D. A., 212 Durlak, J. A., 87
Costanzo, P. R., 77 Dweck, C. S., 60, 168, 182, 185, 187, 189,
Covington, M. V., 53, 55, 57, 192, 280 192, 226, 229, 233, 279-280, 290
Cox, P. D., 164, 167,171,255 DyckJ. L., 129, 136
Crandall,V.J.,4,231,253
Crespo, M, 316 Eagly, A., 252, 256, 258
Crick, N. R., 69, 82, 99-101, 105,-106, 108, East, P. L., 289
116, 184, 189, 191,206 Easton, D., 203
Crowley, C, 287 Eccles, J. S., 3-4, 12-15, 57, 60, 77, 134,
Csikszentmihalyi, M, 264, 268, 282, 319 188, 226, 236-237, 241, 279, 318, 328,
Curran, A., 59 334
Cutrona, C. E., 73 Eckert, P., 49, 323
Eder, D. E., 79-80,
Damon, W., 126-127,257 Eicher,S.A.,78,251,253,258
Dandes, S. K., 87 Eisenberg, R, 127, 129,136
Daniels, D. H., 280 Ekstrom, R. B., 314, 316, 326, 328-331
Darby, B., 46 Ellicott, A., 74
Darlington, R., 199 Elliot, D. S., 72, 325, 326, 331-332
Davidson, E. S., 165 Elliott, E. S., 60, 168, 182, 186-187, 192,
Davies, M., 259 279-280
Davis, W. E., 313 Elliott, G., 127
De Groot, E. V., 280 Emerson, E. S., 203
DeBaryske, B. D., 280, 303 Emmons, R. A., 131
deCharms, R., 4, 12,317 Employment and Immigration Canada, 313
Deci, E. L., 12, 229, 280, 317-318 Ensminger, M. E., 314-315, 331
Delgado-Gaitan, C, 328-329 Entwistle,N. J.,229,231
DePaulo,B. M., 102 Epperson, D. C, 236
DeRosier, M. E., 69, 70, 81 Epstein, J. L., 3, 72,78, 80, 84, 89, 248, 252,
deUrioste, R., 212 257-258, 265
Deutsch, M, 257 Erdley, C. A., 99,107, 109-112, 120, 185,
Diefenderfer, K. K., 167 186, 189, 191-192
Diener,C, 185-187 Erikson, E. H., 322
Dipboye, R. L., 173 Evans, E. G., 262
Dishion, T. J., 72, 280, 290, 320
Ditner, E.,317 Fagan, J., 325
Dodge, K. A., 47, 66, 69, 82, 98-104, 106, Fan, L., 280
111, 120, 181, 183-184, 186, 191,205, Farley, F. H., 79-81
207, 212, 216-217, 233, 281, 303, Farrell,E., 316, 328-329, 332
322-323, 353, 357 Farver, J., 292
Dohn, H., 325 Fein, G. G., 68
Donaldson, G. A., 313 Feiring, C, 282, 290
Dornbusch, S. M, 241, 319, 330 Feldhusen,J. F.,219,228
Doubleday, C, 57 Feldlaufer, H., 236-237, 279
Downs, W. R., 250 Feldman, E., 99, 120
Dowrick, P. W., 165 Feldman, L., 127, 129, 136, 141
Dreeben, R., 235 Feldman, N. S., 259
Dryfoos,J. G.,314 Feldman, S. S., 127,227
DuBois, D. L., 81,262 Felner, R. D., 262
Dubow, E. G., 249, 261-262, 266 Felson, R. B., 25
Dubro, A. F., 269 Figuroa-Munoz, A., 45, 49
Dumas-Hines, F., 186, 192 Finck, D. N.,231
364 Author index
Fine, M, 316 Goodnow, J. J., 233-234

Finn, J., 43, 324, 327, 331-332, 336 Goodyer, I. M, 71
Finn, M., 218 Gordon, L. V., 106, 184, 189, 191
Fiore, M., 66 Gordon, T., 59
Firestone, W. A., 318 Gore, S., 66
Fischer, K., 358 Gottesman, R. L., 283
Fitzgerald, P. D., 103-104 Gottfried, A. E., 280
Flanagan, C , 318 Gottfried, A. W., 280
Fleming, J. S., 280 Gottfried, E., 13
Flink, C , 318 Gottman, J. M, 287
Florido, S., 235 Graham, S., 44, 47, 52, 57-58, 102, 104, 118,
Folkes, V. S., 45, 49 121,279,351,352-353,355
Ford, D. H., 128-129, 135, 148 Green, K.D., 206, 281, 320
Ford, M. E., 79, 126-132, 134, 136-138, 141, Green, R. W., 69
143, 145-146, 148-149, 184,227-230, Greene, D., 12
233, 242 Greene, R. W., 320
Fordham, S., 54 Griesler, P., 70
Forehand, R., 206, 281, 320 Griffitt,C, 129, 136
Fox, N. A., 82 Grolnick, W. S., 279-280, 317-318, 329
Frame, C. L., 102-104, 111, 120, 186 Gronlund, N., 320
France-Kaatrude, A., 165 Grotpeter, J. K., 69, 116
French, D. C , 70 Grusec, J. E., 233-234
Frese, W., 314,330 Guarino, P. A., 57
Frijda, N. H., 135 Guerra, N. G., 82, 102, 106, 110-111, 119,
Fromm-Reichmann, F., 73 185
Furman, W., 68, 70, 76-78, 203, 209, 236, Gunn,T. P., 167
238, 248, 260, 264, 266-267, 282
Futterman, R., 226 Hahn, A., 313-314
Hall, W. M., 291
Gaelick, K., 157, 170 Hallinan, M. T., 79-80, 254, 262, 270,
Galaskiewicz, J., 284, 288 281-282,289-290
Galvin, P., 252, 256 Hamilton, C. E., 207-208, 212, 217
Gariepy, J. L., 71, 284-285, 287, 290 Hamilton, L. B., 279
Gastright, J. F., 324 Hamilton, V. L., 235
Gauze, C , 91 Hamm, N., 252, 256
Gavin, L. A., 266-267 Hammen, C , 74
Ge, F., 280 Handel, S., 45-46
Gerard, H. B., 257 Hanner, T. J., 32
Gershman, E. S., 203 Hansell, S., 80, 267
Gershoni, R., 79, 264 Hanson, A. R., 164-165, 171, 255
Gest, S. D., 71,285,290 Harari, O., 53
Giammarino, M., 290 Hargreaves, D. H., 235-236
Gilligan, C , 32-34, 37,39 Harlow, R., 131
Gitlin, M., 74 Harold, R., 318
Goertz, M. E., 314, 316, 326, 328-331 Hart, C. H., 201
Goetz,T. S., 186, 187,290 Hart, S. S., 279
Goff, S. B., 226 Harter, S., 12-13,15, 17,25,28-31,33,37,
Goffman, E., 49, 60 59-60, 328-329, 334
Goldman, J. A., 212 Hartup, W. H., 3, 45, 68, 83, 216, 227,
Goiter, B. S., 201 251-253, 255, 258, 260-261, 264, 280,
Gonzales, R., 33 319
Goodenow, C , 237, 318 Harvey, O. J., 291
Author index 365
Harwood, R. U 270 Jagacinski, C. M, 55, 57

Haughwout, P., 316 James, W., 24-25
Hawkins,! A., 261, 264, 269 Jamison, K., 74
Hayden, B., 102 Jeffery, R. W., 156
Hayes, D. S., 203 Johnson, D. W., 279
Hazier, R. J., 207 Johnson, E., 35
Hendricks, B., 314, 321 Johnson, J. H., 67
Henkins, A., 287 Johnson, R., 279
Herman, C. P., 252 Johnson, T., 279
Hess, R. D., 280 Jones, E. E., 49-50
Hester, S. K., 235-236 Jones, R. T., 66, 73
Higgins, A., 337 Jones, W. H., 73
Higgins, E. T., 75, 83, 86, 184 Jordan, J., 37-38
Higgins, R. L., 45-47, 352 Julin, J. A., 290
Hinde, R. A., 203 Jussim, L., 283
Hinojosa, D., 326, 328 Juvonen, J., 43, 4 5 ^ 6 , 50-51, 53, 55-56,
Hirsch, B. J., 81 58-59, 252-253, 256-356
Hoehn, S., 353
Hoffman, M. L., 129, 136 Kaczala, C. M., 226
Hofman, J. E., 79, 264 Kakihara, C , 45, 49
Holloway, S. D., 280 Kandel, D. B., 72, 89, 252, 255, 259, 265,
Holm, R., 287 281,290-292,302,327,331
Holmlund, W., 320 Kanfer, F. H., 157, 170
Holt, K., 280 Kaplan, A. G., 38
Holzman, T. G., 167 Kaplan, M. G., 292
Hood, W. R., 291 Karpathian, M., 280
Hoover, J., 207 Katkovsky, W., 253
Hopps, H., 228 Katz, P. A., 279, 318
Horn, W. F., 227 Kearney, C. A., 85
Hoving, K. L., 252, 256 Keefe, K., 60, 168, 175, 248-249, 251-252,
Howell, F. M.,314, 330 261-262,265-267,270
Howes, C , 203-204, 207-208, 212, 217, 292 Keller, J., 279
Hoyle, S. G.,261,264,269 Kelly, D. H., 326, 333
Hoza, B., 70, 80, 202-203, 260 Kelman, H. C , 254
Hryshko, A., 249, 261-262, 266 Kindermann, T. A., 71, 280, 284, 287, 290,
Hudley, C , 102, 104, 118, 121, 353
293, 304, 329
Huizinga, D., 72
Kinney, D. A., 322
Hunt, P., 79-81
Kishor,N.,321
Hyman, C , 69
Kisnadwala, H., 85
Hymel, S., 48, 60, 98, 205, 284, 317-318, Kleefeld,C, 174
324 Klein, R., 83, 86
Knight G. P., 269
Iannotti, R., 129 Kochanska, G., 150
Isenberg, D. J., 257-258, 268 Kochenderfer, B. J., 199, 202-204, 207
Ito, S., 136 Koeppl, G. K., 98, 118,267
Kofkin, J., 37
Jaccard, J., 236 Kohlberg, L., 66, 337
Jacklin, C , 80 Koslowski, B., 12
Jackson, B., 12 Kowalski, P., 15, 17
Jackson, P. W., 235 Kowalski, R. M., 49-50
Jackson, S., 290 Kozeki, B.,229,231
Jacobson, L. P., 289 Kranzler, G. D., 326
366 Author index
Krasnor, L. R., 99 Lynch, M, 208

Krebs, D. L., 129 Lyons, N., 32
Krehbiel,G., 98,218,228
Krumboltz, J. D., 126-128 Macan, T. R, 173
Kuhlen, R.,320, 331 Maccoby, E. E., 3, 80, 146, 150, 227,
Kukla, A., 50, 185 233-234, 252
Kupers, C. J., 164 Mace, F. C, 158
Kupersmidt, J. B., 66, 69-70, 72, 81-82, Maclver, D., 318
98-99, 106, 183, 217, 321, 323, 331, 333 Maehr, M. L., 4, 57, 229
Kurdek, L. A.,261 Magnusson, D., 292
Main, D. S., 318
La Greca, A. M., 87,217 Manni,J.L., 79-81
LaCrosse, J., 66 Marcia, J. E., 322
Ladd, G. W., 82, 98, 105-106, 108, 117, 184, Markus, R, 131
189, 191, 199-207,209-212,214, Marold, D. B.,25,31,37
217-218, 236, 267, 281, 284, 304, Marsh, H. W., 259
318-319 Marshall, RR., 212
Lahey, B. B.,217 Martin, J. A., 227, 233-234, 252
Lambert, N.M., 228, 320, 331 Martinez-Pons, M., 170-171
Landesman, S., 236 Maslow, A., 347
Larivee, S., 162 Masters, J. C, 171
Larson, R., 264, 268, 281, 319 Matheson, C. C, 207-208, 212
Latham, G. P., 168 McCandless, B. R.,212
Laursen, B., 249, 261, 264 McCarthy, K. A., 259, 318
Laychak, A. E., 72, 257, 270, 281, 291, 329 McCaul, E.J.,313,325,328
Lazar, I., 199 McClaskey, C. L., 120
Leal, L., 279 McCleary, K., 250
Leary, M. R., 3, 49-50, 347 McClelland, D. C, 4, 347
Lee, S., 280 McDougall,P.,318
Leggett, E. L., 182, 186-187, 189, 226, 229, Mclver, D. J., 280
233 McKeachie, W. J., 4
Leippe, M. R., 252 McKillop, K. J., 60
Lepper, M. R., 12, 170 McMaster, M. R., 129, 136
McWilliam, R. A., 292
Lerner, J. V., 289
Mead, G. R, 24-25, 27
Lerner, R. M., 289
Meece,J. L., 134,226,235,280
Lewis, M., 282, 290
Meichenbaum, D., 160, 167, 313
Liberman, A., 252, 256, 258
Mellor, F. J., 235, 236
Lindsey, R., 328-329
Mensch, B. S., 327, 331
Littman, R. A., 88
Mettetal, G., 53
Lloyd, D.N., 314-315, 330
Michelena,J.,316
Lochman, J. E., 69, 82, 107, 111, 191-192,
Midgley, C, 4, 12-15, 57, 60, 188, 226,
267
Locke, E. A., 168 236-237, 241, 279, 318, 328, 334
Loeber, R., 72 Miller, B. A., 292
Loebl, J. R, 259 Miller, D. L., 118
Loomis, C.C., 186, 192 Miller, J. B., 38
Lord, S. E.,318 Miller, J. G., 270
Love, R., 136 Miller, K. A., 251-252, 249
Lummis, M., 280 Miller, L., 326, 328
Lundgren, S., 268 Miller, N., 236
Lyman, S., 47 Miller, P., 129, 136
Lynch, J. H., 318 Mills, R. S. L., 82
Author index 367
Mink, O. G., 326 Parker, J. B., 320

Mischel, W., 254 Parker, J. G., 66, 70-72, 78, 80, 83, 98,
Mize, J., 267 203-204, 206, 209, 214-215, 228, 260,
Moely, B. E., 279 266,281,319-320,331
Monsour, A., 30 Parkhurst, J. T., 84, 87, 98, 181, 183-184,
Moore, A. M., 259 324
Moore, S., 212 Pascarella, E. T., 333
Moreno, J. L., 288 Patashnick, M., 53, 263, 268
Morgan, M., 169 Patrick, B. C , 279, 283
Mount, M. K., 174 Pattee, L., 281, 284
Mueller, E. C , 68, 203 Patterson, C. P., 69, 70, 72, 170
Murdock, T. B., 50, 51, 53, 55-56, 59, 253, Patterson, G. R., 72, 81, 88, 280, 290, 303
256, 356 Peguero, G., 328-329
Mussen, P. H., 127 Peplau, L. A., 73
Perlman, D., 73
Nakamura,C. Y.,231 Perrin, J. E., 285
Nasby, W., 102 Perry, D. G., 82, 107-110, 121, 185, 191
Neckerman, H. J., 71, 282, 285, 289-290, Perry, L. C , 82, 107-110, 121, 185, 191
314,321,331,333 Perry, T. B., 78, 203, 209, 259, 261, 264
Neeper, R.,217 Pervin, L. A., 104, 181,229
Newcomb, A. F., 80, 203, 281, 284, 303 Peter, N., 50
Newman, J. P., 102, 111 Peter, O., 352
Newman, R. S., 172-173 Petrie, S., 78
Nicholls, J. G., 14, 53, 55, 57, 226, 229, 263, Pettit, G. S., 120
268, 280 Pettitt, L., 37
Nichols, C. W., 126-128, 131-132, 138, 229 Peyronnin, K., 290
Nisbett, R. E., 12,249 Phelps, E., 257
Noddings, N., 127, 143 Phillips, A. P., 173
Nolen, S. B., 263, 268 Phillips, D. A., 212
Norman, N., 60 Piaget,J.,248,252,259,261
Pianta, R. C , 208-210
O'Brien, S. F., 72 Pierce, G. R., 262, 270
O'Sullivan, R. G., 316 Pink, W. T., 326, 333
Oakes, J., 265 Pintrich, P. R., 280
Oden, S., 117, 119, 187, 189 Pittman, R. B., 316, 323-328, 332-333
Oetting, E. R., 88 Pittman, T. S., 49-50
Ogbu, J. U., 54, 240 Pollack, J. M., 314, 326, 328-331
Ogilvie, D. M., 87 Pollina,S., 136
Ohannessian, C. M., 289 Powell, A. L., 129, 136
Ollendick, T. H., 69, 79, 320 Powelson, C. L., 201,313, 317-318, 327,
Olson, J. M., 252 335,338
Omelich, C. L., 57, 192 Pratt, D., 292
Ouellette, J. A., 268 Preston, A., 253
Price, J. M., 98, 200, 202, 236, 281
Pabon, E., 325 Prinz, R. J.,217
Packard, T., 227 Profilet, S. M.,201
Pandey, J., 52 Putallaz, M., 261,267
Parad, H. W., 290
Parent, S., 162 Rabiner, D. L , 106, 184, 189, 191, 206, 213
Park, K., 72, 82, 251-252, 257, 270, 281, Radke-Yarrow, M., 127
291,329 Rao, N., 279
Parke, R. D., 116 Rasmussen, P., 108, 109-110, 121, 185
368 Author index
Rebane, K., 337-338 Schunk, D. H., 155, 157-171, 174, 176, 255,
Reid, G., 249, 261-262, 266 280
Reid, M., 236 Schwager, M. T., 172
Renshaw, P. D., 75, 82, 102, 105, 117, 187, Schwartz, A. J., 318
191,261,263 Schwartz, F., 254-256, 259
Reuman, D., 318 Scott, M, 47
Rice, J. M., 167 Sebald, H.,251
Rice, W. K., 316 Sedikides, C , 75
Richards, M.H., 281, 319 Seidel, J. F., 325
Rickard, K., 320 Seltzer, V. C , 319
Ricks, D., 66 Sessa, V. I., 290
Robins, C. J., 74 Sewell,T.E., 79-81
Robins, L. N., 314 Shahani,C, 173
Robinson, N., 29 Sharabany, R., 79, 264
Rock, D. A., 283, 314, 316, 326, 328-331 Shaver, P., 73
Rockhill, C. M , 112 Shaw, K., 87
Rohrkemper, M. M , 280 Shea, M. C , 158
Rokeach, M., 269 Sheinman, L., 318
Rollins, K., 208, 210 Sheppard, B. H.,261,267
Rook, K. S., 71,73 Sherif,C. W.,291
Rosenberg, F., 13 Sherif, M.,291
Rosenberg, M., 13,30 Sherman, L. W., 80
Rosenblum, S., 318 Shultz, J., 235
Rosenthal, T. L., 154-156, 163 Shyu, S. J., 258
Rubin, K. H., 68, 82, 99, 217, 284 Sieber, R. T., 236
Ruble, D. N., 19,259 Siesfeld, G. A., 129, 136-137, 141,143, 184,
Rumberger, R. W., 313-314, 324, 326, 242
328-329 Silverberg, S. B., 53, 77, 250, 258
Russell, D., 73 Silvern, L., 318
Rutter, R. A., 314, 325, 328 Simmons, R. G., 13,57,60
Ruvolo, A., 131 Sinclair, R. J., 261
Ryan, R. M, 12, 201, 229, 279-280, 313, Singleton, L. C , 205
317-319,327,329,335,338 Skinner, E. A., 53, 183, 209, 212, 235, 279,
281-283,295-296,304
Sackett, G. P., 287 Skinner, M. L., 290
Sadker, D., 35 Skowronski, J. J., 75
Sadker, M, 35 Slaby, R. G., 82, 102, 106, 110-111, 116,
Safer, D. J., 228 119,185
Sagotsky, G., 170 Slavin, R. E., 134,267
Salomon, G., 162 Slusarick, A. L., 314-315,331
Sancilio, M. F.,319 Smetana, J. G., 242
Santrock, J. W., 171 Smith, H. W., 259
Santulli, K. A., 279 Smith, J. E., 326
Sarason, B. R., 262, 270 Smith, M, 71,256,281,329
Sarason, I. G., 262, 270 Smith, W. P., 165
Savin-Williams, R. C , 261-262, 264 SmollarJ., 53,236,261
Schadler, M., 45 Snyder, C. R., 45-46, 57, 352
Schiefele, U., 280 Solano, C , 73
Schlenker, B. R., 46, 50, 60 Soni, R. T., 289
Schmidt, C. R., 79 Spencer, M. B., 280
Schonert-Reichl, K. A., 337-338 Srebnik, D. S., 270
Schultz, E. M.,316 Stabb, S. D., 118
Author index 369
Stanowicz, L. B., 79 Valsiner, J., 304

Statistics Canada, 313-316, 321, 326-331 Valverde, S., 323, 325
Steinberg, L., 53, 77, 241, 250, 258, 319, Vandell, D. L., 68, 203
330-331 Vandenberg, B., 68
Steinberg, M , 208-210 Vaughn, S., 325
Stevens, E., 129, 137, 141, 143, 184,242 Verette, J., 45, 49
Stevenson, H. W., 280 Veroff, J., 4, 229, 231, 255-256, 259
Stewart, S. L., 82 von Eye, A., 287
Stigler, J., 280 Vosk, B., 206, 281, 320, 331
Stiller, J. D., 318 Voss, H. L., 325-326, 332
Stipek, D. J., 279-280
Stiver, I. P., 38 Waas, G. A., 70
Stocker, C , 29 Wagner, E., 47, 60
Stokes, J. P., 73 Walbek,N. H., 175
Stone, W. U 87 Walchli, S. B., 266
Stoolmiller, M., 290 Wall, S., 208
Stoup, A. L., 314 Walters, H. E., 326
Strauman, T. J., 83, 86 Walters, R. H., 154,252
Stucky, R. J., 45 Wasserman, S., 284, 288
Sullivan, H. S., 73, 87, 248, 259, 263, 269 Waters, E., 208
Surrey, J. L., 38 Waters, P., 33, 37
Suttles,G. D.,251 Wayland, K. K., 82, 107, 111, 191
Swartz,C. W., 165, 168, 170-171 Weary, G., 49
Wehlage, G. G., 314, 325,328
Weinberger, D. A., 227
Takanishi, R.,315
Weiner, B., 44-47, 49-50, 54, 58, 185, 187,
Tamplin, A., 203
226, 249, 252, 256, 269, 279, 281, 347,
Tangney, J., 358
349-350, 352, 355
Taylor, A. R., 105, 182, 184
Weiss, R. S., 73, 76, 209
Tedeschi, J. T., 60
Weist M. D., 69, 320
Terrill, F. J., 174
Weisz, J. R., 279
Terry, R., 69
Wellborn, J. G., 4, 131, 201, 215, 229-230,
Tesser, A., 71, 173, 256, 260, 281, 329
279-283, 287, 295
Thomas, R. J., 326
Wellman, B., 284, 288
Thompson, D. N., 167
Wentzel, K. R., 1, 43, 69, 79-81, 99, 105,
ThurstonJ. R., 219, 228
111, 126-129, 131, 136-137 141, 143,
Tidwell, R., 325
145-146, 149, 181, 183-184,211,217,
Tinsley, B. R., 205
219, 226-227, 229-232, 234-235, 237,
Tinto, V., 326, 332
242, 249, 263, 266-268, 280, 303
TisakJ., 249, 261-262, 266
Wessels, K., 235
Titmus, G., 203
White, B.J., 291
Toles, R.,316
White, K. J., 82, 107, 111, 191
Treder, R., 236
White, R.W., 317, 328-329
Triandis, H. C , 269-270
Whitebrook, M.,212
Tseng, M. S., 326
Whitesell, N. R., 15, 17,28,31,37
Tudge, J. R. H., 257
Wick, P., 87
Tuma, N. B., 80
Wigfield, A., 77, 134,280,318
Turnbull, W., 49-50
Williams, E., 102, 104,353
Williams, G. A., 98, 324
Ullmann, C. A., 320 Williams, J. G., 73
Updegraff, R.,212 Williams, R. A., 281-282, 289-290
Urberg, K. A., 258, 292 Williams, S. B., 325
370 Author index
Wilson, T. D., 249 Zahavi, S., 121

Wood, D. N., 129, 136-137, 141, 143, 184, Zahn-Waxler, C , 127-129
242 Zanna, M. P., 252
Wood, W., 268 Zaslow, M. J., 315
Worthman, C , 50 Zimbardo, P. G., 252
Wright, C , 71 Zimmerman, B. J., 154-156, 159, 161-163,
Wright, J. C , 290 170-172, 174, 176
Zinck, B., 317
Young, J. G., 85 Zmuidinas, M., 352
Youniss,J.,53,236,261
Subject index
Academic skills and goals, 185, 186, 191

and social influence, 156 and intervention of aggression, 118
and self-regulation, 156 for exam, 52
Achievement luck, 52
and social approval, 354 of ability, 50-53
goals and feedback, 166 of controllability, 47, 48
strivings and social relations, 348 of effort, 50-57
and self-presentation 59-61 of hostility, 103, 120
Affect. See also Emotion of intent, 47, 100, 102, 104, 113, 186, 353
as motive, 84 theory, 5, 44, 57, 58, 349, 352
dejection-related, 83-85 Authoritative parenting, 241
cues, 351 Autonomy, 317, 318
Agency beliefs, 130, 139
and academic difficulties, 98 Belongingness, 131, 145, 281, 326
and aggression, 98 need of, 348
and caring, 140
and emotions, 135 Caring, 237
and values, 134 and peer relationships, 203
development of, 149 in schools, 128
Aggression, 218, 219 Caring competence, 136-137
and peer acceptance, 6, 117 and achievement, 145
and school adjustment, 98-99 individual differences in, 144, 146, 147,
and social skills, 117 150
interventions, 117-122 and resource provisions, 140-141
legitimacy of, 100, 110, 111, 119 sex differences, 143
reactions to, 353 Closeness, 210, 211, 213. See also Intimacy
sex differences, 116 Competence beliefs, 14-21, 25, 26, 38
Alienation, 325 and goals, 185,
Anxiety, 210, 219 feelings of, 317
Approval. See also Social acceptance, Sup- Competing goals, 130
port Competition, 263
by parents, 27, 39, 51 and friendships, 256, 262
by peers, 27-30, 38, 51 and motivation, 57, 60
by teacher, 12,27,38-39,51 and self-esteem, 261
Aspirations, 25. See also Expectancy Conflict, 45, 46, 71, 264, 267
Attachment, 208, 209, 217 and peer rejection, 220
Attribution, 351,356 teacher-student, 211,212,214
371
372 Subject index
Conformity, 254, 258 Family

Cooperation, 230, 236 context and motivation, 280
and achievement, 267 socialization, 4
Crowd membership, 78 Friendship(s), 202, 203, 284
and academic tracking, 265
and achievement aspirations, 282
Dependency, 211-214
and achievement, 264
Depression, 73, 74, 85
and conflicts, 263
Dropout(s), 219
and disruptive behavior, 265
academic difficulties, 313-315, 332, 333
and maladjustment, 70, 71
aggression, 331, 333
and school adjustment, 268, 271
and extracurricular activities, 326
and similarities, 282
and social climate, 316
assumptions about, 206, 252
antisocial behaviors, 331
cultural variations in, 270
attitudes toward school, 316
influence of, 248, 250-252, 255, 281, 347
early rejection of, 336
lack of, 70
familial factors, 313-315, 330
longitudinal effects of, 262
peer influence, 327, 337
motives for, 249-252, 259, 271
rates, 313
quality, 203, 204, 248, 249, 259, 266, 263,
risk factors, 313, 337
271
social affiliations, 327
selection of, 254, 281
sociometric status, 320, 323
voice, 315
Goal(s)
academic, 182-183,229,232
Educational environment, 201, 219 and self-esteem, 184
and motivation, 17-19, 22, 24, 57, 60, 188, coordination, 181, 183, 189, 190,218,229,
279 242
and relationships, 202, 214-216 defensive, 192
Emotion. See also Affect dominance, 191
anger, 44-47, 104, 112, 350, 351 hierarchy, 192
empathy, 129, 140-141, 143 human goals, taxonomy, 132
guilt, 350,351 individualistic, 261
other-directed, 350 learning/task, 186,229
pity, 350, 351 performance/ego-involved, 186, 192,229
self-directed, 349 personal goals, 138, 146
shame, 350, 351 prosocial, 48, 119, 130, 190, 232
sympathy, 44 pursuit and strategies, 187
Emotional arousal, 135, 136, 149 relationship enhancing, 356
and caring, 139-141 self-protective, 192,357
negative, 140-141 setting and feedback, 169
positive, 140-141 setting and social comparison, 169
sex differences in, 143 social, 182, 183
social functions of, 135 value, 184
Engagement, 282, 283, 293, 298 Group affiliation, 78
and age differences, 296, 297
and peer relations, 318 Help seeking
Excuses, 355 and academic achievement, 171
and approval, 45 individual differences in, 172
and emotions, 46 and models, 172
and aggressive youth, 47 Helplessness, 37, 39
Expectancy, 358. See also Aspirations Homophily, 89, 289, 290, 301. See also Peer
Expectancy-value theory, 4 networks, similarity
Subject index 373
Hopelessness, 37 and cognitive modeling, 160, 163
Identification, 254 Parent-child interactions, 233

Identity development, 32, 33, 60 Peer acceptance, 26, 99, 205, 214, 220. See
and peer cliques, 322 also Peer approval
and voice, 32, 33 and adjustment, 320
Implicit theories individual differences, 105
and goals, 186, 187 Peer affiliation
individual differences in, 189 changes in, 284,300
Impression management, 356. See also Self- and motivation, 287
presentation Peer approval. See also Peer acceptance
and attributions, 48-59 and social skills, 188
Intellectual competence, 226 and substance use, 266
Intentionality, 47. See also Attribution of in- Peer collaboration, 257, 260
tent Peer context and achievement, 280, 283
Internalized speech, 167 Peer group
Intimacy, 76, 264 influence, 72
networks, 282. See also Peer networks
Jr. high school/middle school, 251, 237, 257 process and motivational changes, 303
motivational change during, 14-17, 20, 22, selection, 289-290, 292, 296, 299, 300,
302
35,57 socialization, 290-291, 299.
socialization and motivation, 329
Kindergarten, 211,219 Peer network(s), 7
age differences, 297
Loneliness, 71, 73, 83, 204, 210, 214 complexity, 294
Mixed age classrooms, 292 group socialization, 301, 302
Modeling, 4, 155,236 in adolescence, 294
and cooperative learning, 174 selection of members, 301, 304
and rewards, 160 sex differences, 294
and social skills, 216 similarity, 296, 297
functions of, 159, 160 stability of motivational characteristics,
Models, 164, 165 299
similarity of, 160, 165 stability over time, 298, 300, 305
Modesty 52, 58 Peer popularity, 50, 53
Moral development, 48, 127 Peer pressure, 250-252
Motivation, Peer rejection, 214, 217
developmental differences, 13 and absenteeism, 69
extrinsic, 11-23, 147,334 and academic performance, 206
individual differences, 14-15 and adjustment, 66, 98, 322
intrinsic, 11-23,38, 147,334 and aggression, 69
Motivational Systems Theory, 4, 6, 128-136, and dropping out, 347
148 and social goal pursuit, 106
Motives and submissiveness, 87
altruistic, 269 and loneliness, 206
individualistic, 269 and school avoidance, 206
and dropping out, 7
Need for affiliation, 67. See also Social needs Peer socialization
and antisocial values, 332
Obligation, 145, 146 and educational values, 328
Observational learning, 254, 255. See also Perceptions of competence, see Competence
Social learning beliefs
374 Subject index
Perceptions of responsibility, 45-47, 58, 355 Self-presentation, 45, 49-61. See also Impres-
Prosocial behavior, 129 sion management
and academic achievement, 227 Self-regulation, 6
and emotions, 352 and feedback, 175
and parent nurturing, 234 and goals, 168,242
value of, 240 and internalization, 154
Prosocial development, 127, 129 and parents, 175
and peers, 175
and self-efficacy, 155, 162
Relatedness, 201, 206, 230, 313, 314,
and teaching strategies, 175
317-319
self origins, 166-174
and school participation, 327
social origins, 163-166
Relationship
types, 157, 158
features, 6, 209
Social goals
functions, 68, 215,216
and achievement, 127
motives and values, 269
and obligation, 126
systems, 212-214
and support, 127
Reputation, 48, 60
hostility, 100, 105-107
Responsibility, see Perceptions of responsi-
Social needs
bility
and antisocial behavior, 105, 115
Romantic relationships, 79
developmental differences in, 77
ethnic differences, 81
School individual differences in, 81
attitude toward, 204, 211, 249, 261, 266, instrumental, 106, 116
272 relational, 106
environment, see Educational environment sex differences in, 80
Self-discrepancy, 68 Social responsibility, 126-128, 131, 141,
and achievement, 162 149-151
and aggressive behavior, 108, 109 and academic achievement, 145-146,
and anxiety, 86 227-228,231-232
individual differences in, 88 Social comparison, 13-20, 26, 38, 256
and social influence, 89 and achievement, 228
theory, 84 and standards, 158
Self-efficacy, 6 Social approval, 25, 49, 51-53, 257
and feedback, 161 Social competence, 4, 226, 227, 239, 255
and prosocial behavior, 105, 108, 109 and friendships, 261
and rewards, 159 developmental differences, 259
and social comparisons, 160, 161, 165 socialization of, 234
and withdrawal, 109 Social goals, 75
individual differences in, 162 pursuit of, 230
Self-enhancement, 127 and achievement, 231
and friendships, 256 and aggression, 116, 119-120
Self-esteem, 24-31 dominance, 116
and efficacy, 170, 171 and school adjustment, 2
and friendships, 256, 260 and self-defense, 100, 107
and support, 26—30 sex differences, 112
and values, 25 value of, 107
developmental differences in, 259 coordination, 82
individual differences in, 29-31 Social influence,
Self-evaluation distal, 357
and learning, 172 proximal, 357
and regulation, 171 and family, 101
Subject index 375
and peers, 101 and social network analysis, 303

Social isolation Sociometrics, 281
and adjustment problems, 323-325 Stress and social isolation, 67
Social learning, 254. See also Observational Stressors,
learning interpersonal, 66, 68, 70,
Social motives, 6, 348 relational, 200
and emotions, 354 Support. See also Social support
Social needs, 68, 73, 74 and achievement, 330
and self-discrepancy, 83 and classroom, 239
and sociometric status, 82 and friendship, 204
domains, 74 and teacher, 28, 136, 137,210
types, 76 emotional, 215, 216
value of, 184 from classmates, 28
Social network(s) from friends, 262, 268
analysis, 285, 287, 288 from parents, 28, 39
memberships, 71 from peers, 238
Social reinforcement, 253 instrumental, 215, 216
Social responsibility, 6, 43 relational, 200, 201
and achievement, 131, 146-150,227
and peer acceptance, 232 Teacher
and schools, 126-128 communication, 13
Social rewards, 254 control, 19
Social rules, 43, 47, 58 perceptions, 22
Social skills reliance, 12
and excuses, 48 punishment, 45, 58
intervention for, 118 student interactions, 208
training and motivation, 267 student relations, 201, 207, 211-212, 216,
Social support, 129, 139. See also Support 236, 346
Socialization Time management
classroom behavior, 235 and achievement, 173-174
cognitive skills, 199
family, 241 Victimization, 101, 120,207
models, 233 Voice, 5
Sociometric status, 213,217, 230. See also and gender role orientation, 33-39
Peer acceptance and Peer approval females, 32-38
and adjustment, 321
and social goals, 231 Withdrawal, 217, 353

Jaana Juvonen, Kathryn R. Wentzel, Bernard Weiner

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Jaana Juvonen, Kathryn R. Wentzel, Bernard Weiner

Uploaded by

Copyright:

Available Formats

How are children's social lives at school related to their motivation to

achieve and how do motivational and social processes interact to explain

General Editor: Martin L. Hoffman, New York University

Advisory Board: Robert N. Emde, Willard W. Hartup,

© Cambridge University Press 1996

First published 1996

Library of Congress Cataloging-iti-Pubiication Data

Social motivation : understanding children's school adjustment / Jaana

ISBN 0-521-47324-1 (he)

Transferred to digital printing 2003

1 Introduction: New perspectives on motivation at school 1

Part I Social motivation: Perspectives on self

1 Modeling and self-efficacy influences on children's

Part II Social motivation: Perspectives on relationships

Author index 361

Subject index 371

Thomas J. Berndt Martin E. Ford

Jaana Juvonen Kimberly Schonert-Reichl

Keunho Keefe Dale H. Schunk

Janis B. Kupersmidt Mary Ellen Voegler

GaryW.Ladd Bernard Weiner

incorporating self-related affects and cognitions including pride, probabil-

1. Researchers focused their efforts on the field of achievement

This book begins to redress this imbalance and counter mispercep-

Themes of the volume

self perceptions (Harter), and social cognitions (Erdley; Juvonen; Schunk

Challenges to the field

early 60s, the need for Affiliation, identified by McClelland (1987) as a

tion. Using theories of social cognition and identity development, several

Overview of the chapters

These authors describe the role of peer networks in socializing motiva-

Maccoby, E. E. (1992). Trends in the study of socialization: Is there a Lewinian

constructs will enhance our understanding of individual differences and

to be graded in terms of their relative performance on standardized tests

Patterns of change upon entrance to junior high school

intrinsic motivation would decrease, those whose intrinsic motivation

- Greater emphasis on grades

RE-EVALUATION OF PARALLEL CHANGES IN

- New social reference group

In our longitudinal study we anticipated, therefore, that a certain per-

Students awareness of changes in the educational environment

Figure 2.2. Relationship between changes in perceived competence and

students' perceptions of why students lose their intrinsic motivation. Thus,

Sixth Seventh Eighth

dents felt that teachers emphasized the components of external evaluation,

Evaluation Soc. Irrelevant Feel Stupid

linguistic exchanges (symbolic interactions) with others. Cooley's formu-

The contribution of perceived scholastic competence to self-esteem

theory involves an examination of the correlation between perceived com-

The contribution of teacher and classmate support to self-esteem

examined the correlations between perceived approval from parents,

more highly). Thus, in order to obtain classmate approval, one must be

The directionality of the link between classmate approval

formulation in which the opinions of others are internalized and therefore

Those reporting that their self-esteem precedes approval from others

that teachers observed behavioral ramifications in the classroom. Teach-

through which they can preserve relationships. However, Gilligan's efforts

Lack of voice in the classroom

n Low Classmate Support

matic. Those who endorse a feminine gender role orientation, combined

teachers and students of Flood Middle School for their cooperation in

Gilligan, C, Lyons, N., & Hanmer, T. J. (1989). Making connections. Cambridge,

In this chapter, I will review recent research that my colleagues and I

Broken social contracts: The function of excuses